Eye in the Sky

Nuts.

Basically, I claim below that it is impossible to have a satellite that can hover over one spot on the Earth and also be 200 miles over the surface of the Earth. In reality, these two things are doable in practice, just not simultaneously. I mention Molniya orbits below, which are highly elliptical orbits that dip down as low as 200 miles, but get very high up. At the peak of its orbit, a satellite in a Molniya orbit moves so slowly that it is nearly hovering over one spot. So it is possible to do both in one orbit, and indeed, it is speculated that the satellite which is the topic of this page is in this type of orbit. Also, Afghanistan is within the reach of cameras from this kind of orbit as well.

So, my original premise is incorrect; the satellite can do what the Yahoo!News article says. However, I maintain that the article is a bit misleading (not enough details are given), implying that the satellite can hover and be 200 miles up at the same time. Also, the other things I say on this page are still correct. So, with this caveat, please read on. My thanks to long time Bad Reader Richard Mentock for patiently pointing this out to me.

October 18, 2001

During times of war, perhaps there are better things to worry about than Bad Astronomy. Yet, sometimes it still comes into play.

A Yahoo! News story reports about how the United States has launched a spy satellite to watch over Afghanistan. The U.S. has many such satellites, used to look at many parts of the world. In general, these are satellites that are like Hubble Space Telescope, except they point downward. They can resolve objects only a few centimeters across, and the advantage of this for gathering intelligence is pretty obvious.

The Yahoo! story says two things about the satellite. They say it "... is believed to be a finely focused spy satellite [designed] to hover over Afghanistan." Later in the article, they say "Aviation Week and Space Technology Magazine said Washington had been readying a KH-11 "Keyhole" satellite with imaging resolution down to a few inches from 200 miles (360 km) out in space...".

These two things together - hovering over a target, and being 360 kilometers up - are mutually contradictory. Why?

It's the nature of orbits. An orbit 360 kilometers up means the satellite will circle the Earth roughly every 90 minutes or so. That means, to someone on the ground, that the satellite will move fairly rapidly, crossing the sky in about two or three minutes. That's hardly "hovering". You can actually see satellites yourself; on any given clear night quite a few are visible.

It is possible for a satellite to hover, more or less, over one spot. There is a special class of orbits, called geosynchronous orbits. In these orbits, the satellite goes around the Earth in the same amount of time it takes the Earth to spin around once. A special subgroup of geosynchronous orbits is called geostationary orbits. In that case, the satellite orbits over the Equator, and in general stays pretty much over one spot on the Earth (this is very similar to why the Moon always shows one face to the Earth). To someone on the Earth, the satellite is always near one spot on the sky, appearing for all intents and purposes to hover over one spot.

The broader class of geosynchronous orbits can have the orbits tilted a bit to the plane of the Equator. In other words, sometimes the satellite is a bit north of the Equator, sometimes a bit south (but never very far from the Equator; the orbit cannot be tilted too much or else the satellite will wander off target). To someone on Earth, a satellite in geosynchronous orbit will stay at the same longitude (east-west) direction in the sky, but will move north and south over the orbit. This is not as advantageous for continuous viewing as a geostationary orbit, but does allow the satellite to view areas on the Earth north and south of the Equator a bit better than a geostationary satellite would. Either way, geostationary or geosynchronous, the advantage of these orbits in intelligence gathering is obvious; it's like always having an eye in the sky over one part of the Earth.

In general, the problem is that a geosynchronous orbit means the satellite must be about 40,000 kilometers above the surface of the Earth. That's a long way, and from that vantage point objects look a lot smaller than they would from a much lower orbit. Like most things, there are advantages and disadvantages for both low Earth and geosynchronous orbits. You have to decide which factors are most important when you design the satellite. [Note added November 29, 2001: I said "in general" above because there is a class of orbits called "Molniya" orbits. These are very elliptical orbits, with a 12 hour period. When they are at the top of their orbit (that is, farthest from the Earth) a Molniya satellite moves more slowly, so if the orbit is chosen carefully the satellite can see cities at high latitude for long periods of time. There are lots of webpages about these orbits; try here, or here.]

So the Keyhole satellite (if that's what it was) launched by the U.S. to watch over Afghanistan is either in a low Earth orbit, and moves rapidly, or is in a much higher orbit, and appears stationary. But it cannot be both. Worse, Afghanistan is a bit too far north to get a geosynchronous satellite directly over it. While the country is still visible to a satellite in that orbit, it isn't optimally placed for viewing, and the satellite can never hover over it.

Want to know more about orbits? The Tech museum in San Jose has a very cool page about geosynch orbits, complete with a nifty animated graphic showing how they work. A more technical version (with more detail than I give here) can be found at JPL's Basics of Spaceflight page.