3 ms·
Suppose you have an earthquake at a known location, or you set off a large underground explosion at a known location. This causes several different kinds of wav
by tzs 12y ago
Suppose you have an earthquake at a known location, or you set off a large underground explosion at a known location. This causes several different kinds of waves to radiate out from that location.
Seismometers can detect those waves. How fast the waves of a given type move depends on the kind of rock or soil they are moving through. When we measure how long the wave took to go from the point of origin to a given seismometer, that tells us a little bit about the kind of rock or soil on that wave's path. The wave might have passed through many different kinds of rock or soil, so just looking at one wave doesn't tell much--just that the path had some mix of rock and soil overall that gave that speed.
We aren't limited to one wave, fortunately. Every time something generates detectable waves, we get a little more data on the composition of the rock and soil on the paths between that source and every seismometer that measures those waves.
The more data we get, the more constraints that puts on what kind of rock and soil can be where and still be consistent with all of the observations.
For things reasonably near the surface the deductions from seismic wave analysis can be checked against observation from tunnels and wells to verify that the methods work.