Earth's core far hotter than thought

http://www.bbc.co.uk/news/science-environment-22297915

by groundCode • 13 years ago

54 26 13 years ago

bbc.co.uk

tocomment• 13 years ago

It's a shame there's no good way to harness that.

ovis2• 13 years ago

Although it's admittedly difficult when the higher temperatures are hundreds or thousands of kilometers away, geothermal power plants do in fact harness that.

dkhenry• 13 years ago

If I am not mistaken Iceland uses geothermal to a large extent and is very successful in doing so.

comicjk• 13 years ago

Yes, we could use more research in the technical side of that. Right now hard-rock drilling is so expensive that geothermal is only viable where natural circulation of underground water brings heat up from hotter regions.

blueprint2• 13 years ago

It's important to note the fact that the heat in the core does not exist in isolation of other fields. The Earth is a dynamo system, and the energy in the core is connected to the planet's magnetic field, the energy bound up in its orbit around the sun, and its gravitational field.

3am• 13 years ago

Its rotation produces the magnetic field which keeps the solar wind from stripping away our atmosphere and leaving the Earth a dead planet like Mars. We're getting our money's worth.

aashaykumar92• 13 years ago

This article is fascinating but as someone who doesn't know much about this discipline, I'm hoping for someone to clarify my confusion:

In the article, it states that our core has temperatures comprable to that of the sun. However, the sun is 93 million miles away while our core seems to only be approximately 4000 miles deep. Shouldn't the Earth be hotter than or are there just so many layers that the heat decreases at such a fast rate as it approaches the surface?

blueprint2• 13 years ago

They're comparing the heat of Earth's core to the surface of the sun, not the core of the sun.

ovis2• 13 years ago

The sun heats the earth surface through radiative heating, while the core is well insulated by the mantle and crust. Although they may have similar temperatures, the power output of the sun is certainly larger.

jasallen• 13 years ago

To provide some further context to ovis' (correct) statements, none of our 'heat', or other energy from the sun comes in the form of conduction. It comes in the form of radiation (light, et al) which the sun has in huge, massive, ginormous quantities. In pop culture, we often talk about the heat of the sun as if it's a direct indicator for the amount of power therein, and the amount of power output. But while it's quite hot, its also quite busy showering us with other forms of energy.

Jabbles• 13 years ago

I'm afraid there are many things wrong with your model of how heat works. However a simple factor is that the surface of the sun is ~10000 times larger than the surface of the earth.

monsterix• 13 years ago

> In the article, it states that our core has temperatures comprable to that of the sun. However, the sun is 93 million miles away while our core seems to only be approximately 4000 miles deep.

I was thinking you were about to ask why in the universe are we not able to measure the temperature of earth core that's only 4000 miles away from us and yet we claim to know the temperature of something that 93 million miles away... :)

Someone• 13 years ago

A lot of the radiation that the sun's core emits gets through its surface without interacting with it much.

You can infer that from http://en.wikipedia.org/wiki/Sun:

    Temperature:
    - Center (modeled): 1.57×10^7 K[1]
    - Photosphere (effective): 5,778 K[1]
    - Corona: ~5×10^6 K

=> The sun's surface is about its coolest part. Its core is about 3000 times as hot. That's where the solar heat comes from.

uvdiv• 13 years ago

A lot of the radiation that the sun's core emits gets through its surface without interacting with it much.

This is false. The sun is very opaque; photons from the core could take millions of years to reach the surface.

https://en.wikipedia.org/wiki/Radiation_zone

https://en.wikipedia.org/wiki/Solar_core#Energy_transfer

This is the article you're looking for:

https://en.wikipedia.org/wiki/Corona#Coronal_heating_problem

thrownaway2424• 13 years ago

"Hotter" in terms of temperature, not heat flux. Temperature is just a statistical thing that measures the relative motion of matter. You might read about experiments where matter in a near-vacuum reaches insane temperatures, even hundreds of millions of degrees. In these experiments there is almost no matter involved and it's whizzing past other matter at nearly the speed of light, hence at high temperatures. But there's very little energy involved. The temperature is more of a statistical stunt than anything, really. At some point they stop measuring it in degrees and start measuring it in KeV.

Anyway, you can have very high temperatures without a lot of heat, just the same as you can have extremely high power without a lot of energy. If I drop my phone on a rock, the instantaneous power of the impact might be extremely high, maybe even megawatts, but only for a microsecond. The total energy involved will not be remarkable.

TL;DR: temperature != heat.

Sharlin• 13 years ago

It's about thermal flux (or heat flux) [0] - the rate of heat flow (SI unit: joules per second aka watts) per surface area (SI unit: square meters).

If the Earth's inner core could be duplicated and moved to space so that its apparent size in the sky would equal the Sun's, it would indeed appear to shine just as brightly as our favorite star. However, it would cool and dim very quickly (relatively speaking), all its stored heat being able to freely radiate into the cold space, the ultimate heatsink.

This is because unlike the Sun, the Earth's core has no active energy source to speak of. A part of the heat is generated by the ongoing decay of long-lived radioactive isotopes, but most of it was created billions of years ago as a byproduct of Earth's formation.

So how do Earth's mantle and crust prevent the heat from escaping quickly? Remember the three mechanisms of heat transfer: radiation, conduction, and convection. The core cannot radiate because there's rock in the way. Rock is also a poor conductor of heat. Even though there are convective currents in the mantle, which is malleable and does flow over geological timescales (but is not liquid!) the convection is far too slow to efficiently transfer heat from the core to the surface.

If the core were to suddenly gain a more effective mechanism of generating more heat, it would cause the core temperature to rise. This would, in turn, raise the temperature of the mantle and the crust, until the thermal flux through the surface matched the extra energy created in the core. Similarly, in the core-in-space thought experiment above, the core would cool until the radiated heat was equal to that generated by the radioactive decay.

The attentive reader might have noted that the above implies a relationship between the surface temperature of a body and its radiative heat flux. This is indeed the case: given an idealized black body [1], the Stefan-Boltzmann law [2] states that the total irradiance (radiative heat flux through the whole surface of the body) is proportional to the fourth power of the temperature (in kelvins) and depends on no other variables. Planets, stars, and duplicated planetary cores are not idealized black bodies but can be approximated as such.

That our planet's core is so slow to cool has a few consequences. The convective currents of the liquid outer core generate the planetary magnetic field that protects us from the solar wind and other charged particles. The convection in the mantle, on the other hand, is responsible for volcanism and plate tectonics. Mars, our sister planet in many respects but only one tenth the mass of Earth, lost its youthful warmth much faster. If it ever did have plate tectonics or a global magnetic field, they shut down billions of years ago when the core and mantle cooled.

TLDR: The Sun continuously creates huge amounts of energy that has to go somewhere, the Earth's core a) doesn't and b) is well insulated.

[0] http://en.wikipedia.org/wiki/Heat_flux

[1] http://en.wikipedia.org/wiki/Black_body_radiation

[2] http://en.wikipedia.org/wiki/Stefan%E2%80%93Boltzmann_law

DaniFong• 13 years ago

The temperatures you're comparing are between the core of the earth and the radiation temperature of the surface of the sun. It's way hotter in the sun's core (20 million Kelvin IRRC)

bttf• 13 years ago

Was anyone else led by the title of this post to contemplate the difference in heat between thought and the Earth's core?

bcbrown• 13 years ago

So the previous estimate was 5000 C, and now it's 6000C +/- 500C, and that's "far hotter"? That doesn't seem like a huge adjustment to me.

ecto• 13 years ago

Perhaps a ~17% change is not significant on the scale that we inhabit (20C today depending where you are), but 1000C is a lot of degrees.

jlgreco2• 13 years ago

Indeed. It seems this new estimate puts it into "surface of the sun" territory (5700-8800K).

edwinkite• 13 years ago

Geologist here. Earth's thermal evolution is unsolved decades after we figured out the stars. The Sun's temperature adjusts to energy production and loss on a 10^6 yr timescale (scaling as the time for a photon to diffuse from the heart of the Sun to the cold surface): << the Sun's age. The Earth's temperature adjusts to internal energy production and surface losses on a billion year timescale: of order the Earth's age. Planets have long memories and history matters. The Sun is reasonably well-mixed. Surface spectroscopy probes the make-up of the whole star. Earth is less well-stirred. Seismic imaging of the deep earth maps the edges of vast pods of material, radioactivity unknown, composition unknown (but definitely distinct from the near-surface stuff), age unconstrained but plausibly as old as the planet [1]. Structure and composition matter [2]. It's a hard problem.

But it matters. When you look up at the night sky far from cities, for every star you see there's a habitable-zone Earth-radius planet that's closer [3]. We didn't know that six months ago. We think (for good, but circumstantial reasons) that complex life requires volcano-tectonic resurfacing - necessarily, a hot interior. Given that habitable-zone Earth-radius planets are not in short supply, the difference between fast and slow cooling for planets like Earth is the difference between a Galaxy where most every star system is habitable and one where almost all the planets are cinders.

The core-mantle boundary heat flux Q_CMB estimated in this paper constrains the mantle energy balance

d(E_mantle)/dt ~ Q_CMB - Q_surf + H_radioactive

Surface heat flux Q_surf is ~46 terawatts. Mantle radioactivity H_radioactive is not well constrained but about 10 terawatts [2]. The implication is that despite the high core heat flux, Earth's mantle is cooling fast - maybe 100 microkelvins per century. Volcanism will therefore shut down in much less time than the remaining main sequence lifetime of the Sun. Absent human intervention, the reddening of the Sun won't kill the biosphere, the Earth will.

As the mantle cools, the temperature contrast between the mantle and the core will no longer sustain core convection. Then Earth's magnetic field will power down. Without geo-dynamo shielding against galactic and solar radiation, bad things may happen: the rapid shutdown of Mars' dynamo is one hypothesis for the deterioration of Mars climate ~4 Gyr ago [4]. On the other hand, Earth's magnetic field strength decreased by a factor of 20 during the Laschamp Event ~41000 years ago [5], with no known effects on biology (or human culture).

Diamond-anvil experiments are tough; few grad students make it past quals without breaking a diamond or two. The diamond-anvil technique is hitting diminishing returns, so modest advances are (rightly) celebrated. The same is true for deep-earth seismology and mantle geochemistry. A good new method is mapping the antineutrino flux from Earth. Antineutrinos are produced by radioactive decay and move in a straight line from source to surface. Mapping the Earth with geoneutrino observatories in the deep sea would help determine the power source for plate tectonics [6].

----

[1] Garnero & McNamara: http://mcnamara.asu.edu/Publications/pdfs/Garnero_and_McNama...

[2] Korenaga, "Urey Ratio and The Structure and Evolution of Earth's Mantle", http://people.earth.yale.edu/sites/default/files/korenaga08d... Korenaga is the best mid-career theorist actively working on this problem.

[3] New result, from several teams working independently to analyze the Kepler dataset: Caltech http://arxiv.org/abs/1303.3013 (read this one first); Harvard team #1 http://arxiv.org/abs/1302.1647; Harvard team #2 http://arxiv.org/abs/1301.0842; Berkeley http://arxiv.org/abs/1304.0460. I'm assuming 0.1 stars per cubic parsec.

[4] Lillis et al: http://seismo.berkeley.edu/~manga/lillisetal2008b.pdf. Later work broadly supports his conclusion that the dynamo died fast and early in Mars history. SETI Institute talk: https://www.youtube.com/watch?v=REiKzxWbzrQ It is not known whether loss of the magnetic field had a big or small effect on the Great Drying of Mars. Measuring modern atmosphere/water loss rates from modern Mars is the goal of the MAVEN mission, which launches this Nov - http://lasp.colorado.edu/home/maven/

[5] Known from ice-core spikes in beryllium-10 (isotope produced by cosmic radiation hitting Earth's atmosphere) as well as magnetic paleo-intensity measurements in sediments.

[6] http://www.phys.hawaii.edu/~sdye/hanohano.html. A knuckle is that SSBN reactors also emit neutrinos and neutrinos cannot be shielded, so deep-sea geoneutrino detectors could be strategically destabilizing. In practice either angular resolution or massive size would be needed to make deep-sea neutrino detectors useful to militaries.

Earth's core far hotter than thought

Related Stories