# Average temperature of the earth - volume, not surface

Do we have a decent enough handle on what happens below the mantle that someone might be able to provide me with an average temperature of the Earth? I don’t mean the surface temperature, but the temperature through the whole volume of the thing. I guess this is something like a measure of the thermal energy of the Earth?

They’re making estimates on planets that we can only see because they caused a tiny anomaly in the light from that star. Thus I’m sure there are ones of Earth too - but I think inaccurate ones. What is beyond a few km deep, we only know theoretically. That is to say it is a matter of speculation.

The deeper you go, the hotter it gets. While some of the heat is due to all of that mass being compressed together, a lot of it actually comes from radioactive decay.

From wikipedia, here:

As was already mentioned, a lot of this is theoretical. It’s not like we’ve actually stuck a thermometer down into the earth’s core or anything.

Almost none of it. Just being under pressure doesn’t give you any heat at all. The process of becoming compressed produces heat, but that all happened billions of years ago. There’s been more than enough time since then that, if there were no other heat source, we’d have shed almost all of that original heat by now.

In fairness couldn’t “all of that mass being compressed together” also be considered “residual heat from planetary accretion” per below?

Also note that the normal “rule of thumb” geothermal gradient figures of 10-30ish degrees C per KM only apply to the crust, since the radioactive elements that create the bulk of the earth’s heat are concentrated in the crust. It’s probably more like a degree per kilometer once you get past the Moho. Realistically, for the purposes of this exercise you could probably just ignore the crust since it’s less than 1% of the earth’s total volume. Just call the top of the mantle 700 degrees C or so and then add 1 degree/km all the way down.

I think that’s what I was considering it to mean. Though I misremembered the figure as being much smaller than 20%, so my statements of “almost all” were a bit stronger than they should have been.

And GreasyJack, if anything, wouldn’t the radioactive elements be more concentrated in the core, since they tend to be quite dense?

Cite? I’d expect the bulk of the radioactive elements - uranium, thorium etc - to be at the core because of their density.

Radioactive elements are what are called “lithophiles” which means they prefer to bond with oxygen, which means that they are very easily integrated into silicate minerals and rocks. In contrast, the elements that are relatively depleted at the earth’s crust like iridium and platinum are “siderophiles” which means they preferentially bond or occur with elemental iron. Even though elements like uranium are quite dense, the rocks and minerals in which uranium commonly occurs are far lighter than the tightly-packed elemental iron in the core. See here: http://en.wikipedia.org/wiki/Goldschmidt_classification

Now, granted, there has been a bit of trouble reconciling the geochemically-derived theories of the composition of the mantle and core with geophysical observations. One of the specific problems is that it seems difficult to account for the amount of heat deeper in the earth with most of the radioactive elements tied up in the crust and upper mantle. The idea that there are large amounts of radioactive elements trapped in the lower Earth somewhere, or even a natural nuclear reactor, is intuitively appealing and has been proposed many times, but thus far these proposals have more problems than merely tweaking the current model which has most of the radioactive elements nearer the surface and the heat in the core being mostly residual and frictional rather than directly derived from radioactive decay. The (heavily caveated) analogy my geochem professor used is that the radioactive elements in the crust are sort of like an electric blanket in that they’re not actually heating the core, but they’re greatly slowing the rate at which the center of the earth loses heat to space.

It’s been a while since I took the relevant classes, but a bit of googling doesn’t seem to indicate there’s been any major upheavals in this regard. In fact, one of the problems with trying to find a cite that concisely sums up the issue is that there are so many articles in popular publications about the various radioactive material in the core theories that have come and gone, but not much that actually explains the current conventional wisdom. There’s a lot of higher level class outlines and such that deal with it obliquely and in very technical terms, but here’s a geo-101 online text that says the bit about (some) radioactive elements in the crust explicitly: