|MadSci Network: Earth Sciences|
As you know, we cannot look directly. We must rely on indirect information and on reasoning about that information. There are several general types of information that are helpful.
One important source of data is the waves that travel through the earth from earthquakes. These waves come in two main types, compressional P-waves and shearing s waves. After a big earthquake, we can find places in the world where no S waves were detected, from this we have been able to conclude that the outer part of earth's core is a liquid, because solids transmit S waves, but liquids do not. We can use the size of the area where no S waves occur to calculate the size of the outer core. We can find the inner core in similar but more complex ways, involving the conversion of S waves to P waves and then the refraction of those waves. These days, even more information is learned by using the same sort of computer data processing that physicians use to do CAT scans with x-rays. Geophysicists can collect earthquake waves coming from many different sources into an array of stations and then use the patterns of arrivals of different kinds of waves to do "seismic tomography" of the earth. It is important to remember though, that seismic waves only give data about density and rigidity. Different materials have similar properties and so it can be difficult to decide exactly what's down in the earth based on seismic data alone.
Another important source of information comes from pieces of the interior of the earth that are transported to the surface in volcanic eruptions. Kimberlites and basalts may erupt from sources as deep as 100-150 km in the mantle and bring chunks of the mantle up with them as they rise. These "xenoliths" can then be studied for their composition and structure. Unfortunately, the deepest xenoliths are still coming from the uppermost mantle, so we can't directly sample very much of the earth at all. The good news is that we can study the physical properties of xenoliths in the lab and compare the results from those samples to seismic data from the same area and improve our ability to interpret seismic data.
Also, we can draw inferences based on what we can see and what we think about the origin of the earth. We believe earth formed from a cloud of debris, the remains of which can still be seen in meteorites and asteroids. After collecting many many samples of meteorites that have fallen to earth, we can get a good idea of what the average chemical composition of the earth must have been to begin with. When we combine this with good knowledge of the rocks in the crust and upper mantle and some inferences from the seismic data about the composition of the core, we can perform some subtraction and end up with what the rest of the earth must be made of. We can test these ideas by taking material of that composition into the lab, placing it under high pressure and temperature and comparing it's physical properties to the physical properties we can measure with seismic data. If the properties match, that says the inferred composition is possible and strengthens confidence in our knowledge of the rest of the earth as well.
Finally, it is very important that you do not get the idea that we "know" exactly what the inside of the earth is made of. We have explanations that fit all of the data outlined above, but major new advances in our understanding of the core have been made within the past two years, and a new idea is now emerging and being tested that suggests that the core of the earth may actually contain substantial amounts of hydrogen. The presence of such a light element in a part of the earth that we know is very dense seems unlikely at first, but the very high pressures in the core make some very interesting things go on. These pressures are very difficult to study, though we can squeeze materials between the faces of two diamonds and get pressures that high in the lab. The idea about hydrogen in the core has only been tested by one person and it has not been widely accepted, but it does seem to explain the information we have better than the current model of a metal-only core. Much remains to be learned about the inside of the earth.
Geology and Environmental Science
La Salle University
Try the links in the MadSci Library for more information on Earth Sciences.