From: utzoo!decvax!harpo!eagle!karn Newsgroups: net.space Title: Re: Cosmos crash - (nf) Article-I.D.: eagle.744 Posted: Wed Jan 26 11:56:02 1983 Received: Fri Jan 28 21:10:20 1983 References: fortune.751 The probability estimates were correct. The earth was rotating inside the (relatively fixed) plane of Cosmos's orbit, and the "mean motion" (number of orbits per day) was somewhat over 16. Over just a half day, the "tracks" as drawn on a map get pretty dense. This means that unless you knew exactly the orbit in which it was going to come down that it could do so almost anywhere between 65 deg north and south latitude. It turns out that the percentages for land, ocean, etc, of the earth's surface don't change much if you discount the arctic regions. The best analogy I can come up with for explaining why reentries are so hard to predict is a roulette wheel with the ball as the satellite - the motion of the ball around the race may be easy to predict, but there are too many unknowns involved when the ball "decays" and starts to bounce off the wheel. The best you can do is count up the colors and estimate probabilities. Perhaps this is an example of Russian Roulette? (sorry) Phil