Many people will already be aware of the large number of orbiting satellites to be seen in the hour or two after sunset and before sunrise. There are probably about 400 objects visible to the naked eye out of the 8,000 or so currently in orbit. The brightest regular object is the Russian Space Station Mir (magnitude 0). The US Space shuttles are brighter - as bright as Jupiter at magnitude -2.5 - but not of course in orbit all the time.
Many of the objects seen are the last stage of a launch vehicle and are rotating, and hence vary in brightness as the visible cross-section varies. Some objects like COBE are deliberately spinning.
Occasionally one sees the Sun reflected off a flat mirror-like surface (solar panels for example) and the brightness can be higher. I have seen the Hubble Space Telescope at magnitude -4, probably from a reflection off the solar panels. I have also seen reflections off the flat solar cell covered tops of several hatbox-like satellites.
Calculations show that one square metre of a mirrored surface at a range of 1,000 kilometres should look like a magnitude -7 star. Such events are unpredictable because the exact orientation of the satellite or the solar panels (relative to the observer) is hard to know in advance for several reasons.
The Iridium Satellites
This situation has changed with the launch of satellites by the Iridium Corporation which finished putting together their satellite version of a cellular telephone network in june 1998. The working system consists of eleven satellites per plane in evenly spaced (30 degrees) planes. This is the first of some 9 proposals to clutter low Earth orbit with systems of communications satellites. The desired orbit is circular, 782km high, and inclined 86.4 degrees to the equator. The 72 required satellites have been launched, and though 4 are already duds for one reason or another, most are working fine. The company has been offering services since October 1998.
Soon after the initial launch it came to the notice of amateur satellite observers that one got some very bright reflections from these satellites (magnitude -8), in certain parts of the sky. Paul Maley of Houston Texas, published the first photo of three of them flashing together when these three were still close together after launch. He also obtained the first video of a "flash" and ascertained details about the geometry by personal enquiries.
Since the satellites are rotating once per orbit, so that the orientation relative to the local vertical is always the same, the directions of satellite surfaces are thus known all the time. The main mission antennae (MMA) have a silver-coated teflon surface of area about 1.6 square meters, so it is reasonable to assume they are the surfaces involved.
Predicting Iridium Flashes
Mr Randy John and Mr Rob Matson have written special versions of their satellite tracking programs SKYSAT and SKYMAP to produce "flash" predictions. Both programs calculate the point in the sky that an a given place on the ground is seeing for the three MMA. This position is compared with the solar direction. If the angle is smaller than an arbritary small angle for any of the three mirrors, then a "flash" will occur. Rob Matson's program includes an estimate of the brightness derived from fitting observed brightness versus flash angle. Because no surface is an ideal mirror brightenings do occur at angles more than 0.5 degrees.
The reflections are characterized as peaking at magnitude -6, lasting for about 10 seconds in a dark sky, and satellite movement of some 5 degrees. One needs binoculars to see the satellites ordinarily as they are magnitude 6 to 7. These flashes occur during daytime as well and provide a means whereby you can say you have been able to see a satellite during the daytime with the naked eye. I have already observed such flashes.
The Heavens-Above in Munich, Germany provides a flare prediction service. Because the flare effect is fairly local in nature, the user should provide their best known coordinates in order to observe the flare successfully.
The Iridium Coporation ceased operation of the Satellites on March 18, 2000 due to its insolvency, and the failure to find a buyer. Eventually a buyer was found, and the transaction approved by the appropiate US Bankruptcy court. The constellation is now owned by Iridium LLC. Currently (September 2001) the system is being marketed at Government and industrial users in remote locations, for both voice and digital data use. On November 30 2000 the first Iridium satellite re-entered, Iridium 79 which had failed on launch. Since then its orbit had been shrinking through the normal process of atmospheric drag. The number of failed iridium satellites is now significant. They are characterized by a rotation period of seconds. When near a point in their pass where flaring would occur they CAN give a set of very short (less than 0.25 second) flashes to negative magnitudes. The IRIDFLAR programme has been written so as to ignore the known tumblers.
The Visual Satellite Observers Home Page has several pages covering more details on this story, Paul Maley's photograph and copies of the prediction program, IRIDFLAR, and pointers to Mr. R. John's SKYSAT program.
Rob Matson can be contacted at firstname.lastname@example.org. His IRIDFLAR program is a DOS program. A version for systems without floating point processors is also available. IRIDFLAR will run on the Macintosh under a PC emulator (Virtual PC or Real PC).