I'm not a geologist, so take the following with a grain of salt:
We rely on a combination of different methods from different fields. For example, we can look at the composition of meteorites, which are believed to be representative of the elements making up the rocky planets, and infer from them the composition of the Earth and other planets. We also know enough about stellar nucleosynthesis (creation of different elements inside stars and supernovas) to know that iron and nickel are
very abundant compared to heavier elements and should form a substantial fraction of the non-gaseous material available for planetary formation.
Note that the average density of the Earth compared to the Earth's crust places great constraints on what elements could possibly make up the internal structure of the Earth. Since the density of the Earth as a whole is greater than the density of the crust, this means that heavier elements must be making up a substantial amount of the internal structure of the Earth. This fits with our current models of planetary formation in that when proto-planets are still hot and molten, heavier elements tend to sink inwards, which should lead to a core composed mainly of the heavy elements.
So just based off of the composition of meteorites, known reactions from nuclear fusion, the density of the Earth vs the Earth's crust, and some simple physics regarding buoyancy, we can already say that the Earth's internal structure must be composed of something similar to iron and nickel.
But that doesn't guarantee that the core itself is composed of iron and nickel. Further evidence comes from the aforementioned seismic waves. If, say, the mantle was composed mainly of iron and nickel, instead of silicon and oxygen, seismic waves would behave much differently than we observe. We'd also have massive amounts of molten iron and nickel popping out of volcanoes, but we don't see that either.
The details get very, very complicated. Just try reading
this paper for example. Or
this one. You'll find that unless you've been trained in geology and whatever else may be required, you will have a lot of difficulty understanding the details underlying all of our models. Unfortunately there is no way to teach this in a forum. You'd need to be formally educated (or devote a
substantial amount of personal time) to learn it all.
None of the definitions
found here of the word 'guess' seem to apply. While there may some guessing involved during the initial formation of a theory or model, the entire process evolves from a guess into a (hopefully) sound model or theory based on months or years of work. I see no reason to consider the end product of this process a "guess".
No, I wouldn't say that at all.
This problem originates from a life in construction vs. a life in general.
)
