Scientists Measure Gravity Change From Earthquake 99
Science Daily is reporting that scientists were able to use satellite data to watch changes in the Earth's surface caused by a massive earthquake. These changes had two major measurable effects on the region. The massive uplift in the seafloor changed GPS measurements, and the density of the rock beneath the seafloor changed which produced a detectable change in gravity.
Re:how does this work? (Score:2, Informative)
Re:Top-notch editing (Score:2, Informative)
Perhaps it was an intentional use of Chiasmus [wikipedia.org] with the intention of intensifying or bringing greater attention to an important point.
Re:magnitude of the change (Score:4, Informative)
Meaning it should be in the 0.01 m/s^2 range.
For the simple reason that if it were anymore, the earth would deform to counter that imbalance (molten core, you know).
That, btw, also limits the height of mountains to about 10-12Km on earth (compare to mars, where to lower gravity constant allowed much larger volcanos)
Re:magnitude of the change (Score:5, Informative)
The full paper as well as a very nice layman's introduction in the Perspectives section is in this month's issue of Science. (Sorry - subscription only. But you may be able to find the text on a preprint server. I'm no geologist, but I haven't been able to find it in any of the obvious places.)
Basically, they map out a change of 15 microgals (1 gal = 1 cm/s^2) or around 1.5e-8 of the average gravitational field on the earth.
By comparison, the variation in g with latitude (at constant elevation) is around 0.5 percent, or 300'000 times as much. Variation associated with local geology is around 100 times smaller, but still swamps this earthquake signal.
What's cool about this measurement isn't that they're measuring something big enough to have any effect on humans, but rather that they're able to measure such a tiny effect at all.
There are all sorts of processes going on in the earth and in the oceans that involve movements of comparable amounts of mass: changes in glacier and polar icecaps, ocean-atmosphere gas exchange, deep sea current and temperature changes, movement and depletion of underground water, fast moving magma associated with volcanos, slow tectonic changes, etc. And now it seems like it's also helpful in trying to construct detailed models of an earthquake.
Incidentally, if you were an athlete trying to cash in on lower gravity, you'd be better off training in the Chilean highlands and competing in Puerto Rico - but it still wouldn't help you much, especially compared to biological effects and day to day variation in performance. (http://www.csr.utexas.edu/grace/publications/fac
Re:GPS is relative to exactly where? (Score:5, Informative)
When you get a position fix from the GPS system, you are combining the information from several satellites, each of which is transmitting a signal of the form, "My name is GPS _X_, I sent this message at time _Y_ from the approximate location _Z_" A GPS receiver triangulates a position by calculating the distance to several satellites, using Center-Of-Earth coordinates.
It's true that your position is calculated relative to the satellites, but in order for the satellite to know where it was when it sent the message, there has to be pretty accurate data about its precise orbit, which depends intimately on the shape and mass distribution of the earth ("Geosynchronous" is only approximate), so that the final location in Latitude.-Longitude.-Altitude can be given relative to the center of the earth. A big quake could certainly shift things around enough to alter the orbit, which is probably what these researchers were talking about.
On the other hand, for the purposes of surveying on earth, its certainly conceivable that one could define property lines in relation to the locations of particular GPS antennae, fixed into bedrock or something, and that if those moved, things would be all kerflooey. But that's not new to GPS surveying, since its always been done relative to the location of particular fixed monuments. :)