Earth, a Giant Pinball Machine

An anonymous reader writes "Scientists have long probed Earth's interior by monitoring seismic waves (if earthquakes don't make them, they can be induced with explosives, and one nuke test actually triggered an earthquake!), which reveal the inner structure of the planet. But what if the method is wrong? LiveScience reports on a new study suggesting Earth is like a pinball machine, with sound waves careening around before they get to the surface. What is interpreted as a broad layer change could be nothing more than a localized density variation."

Comments

If this is true.

[Murphy Murph]

If this is true - and their experiment with the slab of aluminum offers decent evidence - why do modern seismological methods work as well as they do?

Money is the prime driver in many forms of research, and nobody has as much money vested in geologic surveys as the oil companies. Why haven't they already discovered this effect?

Question: Do seismic surveys currently employ a Radon transform (like how CAT scans reconstruct a 2D image from 1D projections)? If so, how would this "pinball effect" affect that?

Re:If this is true.

[mopomi]

Yes, they typically employ a Radon transorm during the inversion. Seismology is very much like a CAT scan w.r.t. methodology. The problem described in the paper (http://acoustics.mines.edu/preprints/vanwijklevsh in04.pdf) is that the multiple scattering caused by small heterogenieties near the surface of an object can cause the same sort of signal (to within error) as a multiply layer object. The radon transfer will have just as much of a problem as any other inverse transform simply because the matrix (data) is "bad". That is, the solution is non-unique and without some other a priori knowledge, we wouldn't be able to distinguish different forward models.

Re:If it's wrong, we'll fix it...

[shaitand]

But we've been using the current method for a long time. What happens if we adjust the model instead of changing of our method, resulting in a new model which appears to work for a long time. In the meantime this discovery is determined not to be correct.

Eventually it is almost forgotten, but remembered just well enough that anyone who it occurs to, and checks, will see that someone else already thought of that and the theory was found to be incorrect (since it clashed with our new false model).

Sure someday we would probably figure it out, but how long would it take to rediscover the correct solution? How long would mankind of have been robbed of true knowledge by Occums razor?