Banana Donuts and Finite Frequency Tomography

billiards · Banana Donuts and Finite Frequency Tomography

Do we have anybody who knows any seismology here?

In case you don't know, seismologists have traditionally been using ray theory to unlock seismic data. Ray theory has been borrowed from optics, the study of optical light, which assumes an infinite frequency signal. This approximation is justifiable at the frequencies of optical light, but when signals get down to the lower frequencies of sound generated by earthquakes and explosions, the approximations break down. Therefore seismologists have needed a new theory, which accounts for the finite frequencies of sound, to better understand the true meaning behind their seismic data.

Matt

I've looked at seismology & finite frequency tomography in the past, but the technicalities of it are a bit beyond me. It seems to give some pretty interesting images of the mantle though.

billiards

I must admit that the technicalities are also beyond me, apparently there are only a handful of people ("about 20" according to my old seismology lecturer) in the world who really understand it. It is based on some pretty rigorous physics and is now widely accepted as being wholly superior to ray theory in its accuracy, but remains a lot more difficult to apply to data sets in reality. The theory is inherently stronger than ray theory as we deal with lower and lower frequencies, indeed, at infinite frequency the theory breaks down to become ray theory. Exploration seismologists have up till now ignored finite frequency tomography because they deal with frequencies typically in the range of 4 to 100 Hz (and are most interested in energy at around 50 Hz), it has been suggested that notable differences in the predictions of finite frequencies as opposed to ray theory only come to the fore at frequencies below about 10 Hz (http://geoweb.princeton.edu/peopl...ulty/nolet/Dahlen_for_dummies.pdf). Therefore the theory has had little opportunity to flower as most of the research funds in sesimology are funded by oil companies only interested in crustal structures and not global structures.[/url]

Matt

I just found this. It seems to have big implications on what we have deduced already from tomography, but again it's outside my scope to really understand how big an effect it will have.

billiards

Yeah there's some pretty interesting stuff going on in the lower mantle. I think that there was some debate as to the behaviour of perovskite at lower mantle conditions. Fairly recently (in the last two or three years) researchers have pretty much determined that perovskite undergoes a phase change to what is now technically termed post perovskite. It's a hot topic in geophysics as to the spin state of post-pervoskite, spin state affects the heat conductivity of a material. If we knew about the heat conductivity we would be able to infer stuff about the mechanisms of heat transfer in the lower mantle; could mantle plumes stem from the lower mantle?