Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The core of the earth

  1. Feb 23, 2007 #1
    How do we know the composition of the earth's core? In addition, what is the current model for the generation of Earth's magnetic field?

    I'm curious to find out.
  2. jcsd
  3. Feb 26, 2007 #2
    Short answer: we don't know.
    Slightly longer answer: we guess based on planetary formation theories (derived from astronomy), as well as seismic and gravitational observations (in conjunction with mineral physics).

    I believe that the magnetic field is generated by currents in the outer core, these currents arise by compositional convection whereby light incompatible elements are disassociated from the solidifying iron at the inner core boundary (although I'm about to spend a whole day learning about this in a couple of days so I could probably tell you more then).
  4. Feb 26, 2007 #3
    i saw something on the National Geographic channel once. they measure the speed of seismic wave propagation through the earth and try to match them with the material they already assumed were found in the earth. as for the magnetic field, something to do with the rotation of the metal core of the earth. cause i remember that if that core stops rotating then we lose our magnetic field thus we lose the protection of the atmosphere..blablabla..all Guessing..
  5. Mar 3, 2007 #4


    User Avatar

    I think it is the other way around. The movie The Core got it wrong and it is actually the magnetic field that causes the spinning. Is that right?
  6. Mar 3, 2007 #5
    My (rather basic) understanding of a dynamo is that the moving conducter in a magnetising field causes electrical currents which keeps the magnetic field going, so there needs to be something that keeps the conductor physically moving to keep the dynamo running. The geodynamo is probably driven by the rotation of the inner core, but I'm not sure how important the spin is, I believe that it is the compositionally driven convection currents associated with the rejection of light elements at the inner core boundary where Fe freezes out of solution, that plays the really important part in driving the dynamo.
  7. Mar 8, 2007 #6
    Problems with the Core as the Source of the Earth's Magnetic Field

    The dynamo, which in theory generates the earth's magnetic field, requires a sharp temperature differential at the core boundary (Core material crystallizing generates the sharp temperature differential but ...) The problem known as the "heat flux" problem is that based on thermodynamic calculations the maximum age of the earth's core is 1 billion years. As it is known that the earth's magnetic field has existed for at least 4.2 billion years, there is no mechanism to generate the sharp temperature differential which is required to create the thermal dynamic movement of conductive liquied core which is hypothesized to generate the geomagnetic field, prior to the formation of the earth's core.

    Comment as to theoretical mechanism which is hypothesized to generate the magnetic field :

    The entire core is above the Curie temperature so the geomagnetic field is supposedly generated by self excitation, where something starts a magnetic field in the core (no explanation for what is the starting mechanism) and the magnetic field generates currents in the conductive core which generate a magnetic field which generates additional currents.

    Another problem for the core model is the recent findings of very rapid geomagnetic field variances, which occur at a rate that is not possible if the geomagnetic field is generated in the earth's core. Based on theoretical calculations a geomagnetic reversal should, if the magnetic field is generated in the earth's core take a couple of thousand years to complete.

    Attached are two paper that record massive field directional changes, that are more than 300,000 times faster than the maximum field directional rate.


    From Acton's paper that discusses the African Afar anomaly: "One lava flow has recorded both of the antipodal transitional components, with the two components residing in magnetic minerals with unblocking temperature above and below approx. 500C,... Hence the configuration of the geomagnetic field, appears to have jumped nearly from a north-hemisphere transitional state to a south-hemisphere one during this normal-to-reverse polarity transition."

    From Coe et al's 2002 paper that discusses the Oregon anomaly: "Paleomagnetic results from lava flow recording a geomagnetic polarity reversal at Steens Mountain, Oregon suggest the occurrence of brief episodes of astonishing rapid field changes of six degrees per day. The evidence is large, systematic variations... in a single flow... most simply explained by the hypothesis that the field has changed direction as the the flow cooled."

    Last edited by a moderator: May 2, 2017
  8. Mar 9, 2007 #7
    if the earth's magnetism is due to revolving currrents, then this magnetism is basically something we've seen in motors and in so many other magnets...while these magnets attract only iron, howcome earth's magnetic field attracts every other object...
  9. Mar 9, 2007 #8
    We've determined the different layers and their compositions by using seismic wave data (earthquakes and nuclear explosions), which showed that the outer core is liquid. The circulation of heavy metals in the outer core is supposed to be what causes the magnetic field.
  10. Mar 9, 2007 #9
    Yes there is, the formation of the inner core creates enough entropy to keep the geodynamo going, if it hadn't started to form we wouldn't have a magnetic field anymore.
  11. Mar 9, 2007 #10
    Geomagnetic Field - High Frequency Changes?

    Set aside what causes the geomagnetic field for a moment. What causes the observed high frequency changes in the geomagnetic field? Note the geomagnetic changes correlates with the semi-periodic abrupt climate changes that were discovered at the end to the 20th century. From the attached paper by Stoner, Channel, et al.

    "The observation of synchronous high-frequency (1000 yr) variations in both the filter (proxy records of geomagnetic paleointensity) and the fitrate (cosmogenic isotope flux) suggests that the variations are due to global scale geomagnetic field changes."

    See figure 9 in the paper that shows the geomagnetic field drops also correlates with abrupt climatic changes (for example Heinrich events, the Younger Dryas is an Heinrich event.)


    The paper quoted in what causes Ice ages asserts that macro-climatic changes are cause by changes in the intensity of the geomagnetic field. In addition another paper which should be added to the "What cause ice ages" provides data to support the assertion that Galactic Cosmic Rays (GCR are basically high speed protons) increase when the solar system passes through the spiral arms of the galaxy.

    An increase in GCR has been shown to result in an increase in clouds. An increase in clouds increases the planet's albedo which results in cooling. The long term increase in GCR levels (times when the solar system was moving through the spiral arms) correlates with the ice epochs. (We are in middle of an ice epoch now.)
    Last edited by a moderator: Apr 22, 2017
  12. Mar 10, 2007 #11
    I had a lecture on this not so long ago, from the top of my head I seem to remember it has something to do with spherical harmonics. Taylor columns form, these are currents which kind of spiral around a poloidal path, I believe that the orientation of the Taylor columns (i.e. parallel to Earth's rotation axis) is what causes the field (over the vast majority of geological time) to have its poles approximately at geographic N and S, this is because the magnetic flux is concentrated where convection currents are downwelling.

    Basically we were taught that the field reverses by the same mechanism that sun spots are formed, what happens is patches of 'opposite polarity' flux gets into the 'wrong' hemisphere. This happens by a twisting of toroidal mode flux lines by irregularities in the convection currents, once it is there it may, if amplified sufficiently, lead to a full reversal.

    With regards to the fast change, I cannot really comment as I have not got the time to read those papers carefully, however I either suspect that the palaeomag data is misleading for whatever reason, or perhaps this is a very localized phenomenon. Typically a reversal should take about 3 - 4 thousand years, and there should be a period wherre the intensity of the field drops right down to pretty much zero in between.
  13. Mar 10, 2007 #12
    Geomagnetic Field Unstable after last Reversal? Why?

    Hi Billiards,

    In response to your comment:

    "Basically we were taught that the field reverses by the same mechanism that sun spots are formed, what happens is patches of 'opposite polarity' flux gets into the 'wrong' hemisphere. This happens by a twisting of toroidal mode flux lines by irregularities in the convection currents, once it is there it may, if amplified sufficiently, lead to a full reversal."

    The sun and the earth are quite different. The hypothesis which you quote can not explain the observed semi periodic variance in the geomagnetic field. (i.e. What is cause the semi periodic variance?)

    I have found a brillant unpublished paper that summarizes the problems with "the geomagnetic field originates in the earth's core" hypothesis. Before presenting that author's hypothesis and attempting to solve the puzzle, let's look at what the geomagnetic field mechanism hypothesis must explain.

    There is new data that indicates that the geomagnetic field, spends 20% of its time in a low state (5 to 10 times less than the current geomagnetic field intensity and unstable. This unstable period coincides with the ice age cycle change from 41 kyr to 100 kyr. See the thread "What cause Ice Ages" for details.) The question is what causes the geomagnetic field to drop and to become unstable? Changes happen for a reason.

    From Zang and Gubbin's attached paper "Is the geodynamo Unstable?"

    http://eprints.whiterose.ac.uk/archi.../gubbinsd4.pdf [Broken]

    Recent palaemagnetic studies suggest that excursions of the geomagnetic field, during which the intensity drops by a factor of 5-10... The 'normal' state of the geomagnetic field dominated by an axial dipole, seems to interrupted every 30-100 kyrs: it may not be as stable as we once thought."

    "These important results paint a rather different picture of the long term behavior of the field from the conventional one of a steady dipole reversing at random intervals; instead, the field appears to spend up to 20% of its time in a weak, non-dipole state (Lund et al 1998). ... This raises a number of important but difficult questions for geodynamo theory. How can the geomagnetic field change, so rapidly and dramatically? Can slight variations of the geomagnetic field affect the dynamics of core convection significantly? If so, is the geomagnetic process intrinsically unstable?
    Last edited by a moderator: May 2, 2017
  14. Mar 10, 2007 #13


    User Avatar
    Gold Member

    Um. Like what? You're not confusing the magnetic field with gravity are you?
  15. Mar 10, 2007 #14
    Okay, I'm slowly returning online after a large move and I'm not following this discussion, but that statement triggered something. WWe know that Heinrich event 1, the last one, timewise, is dated around 17-16,000 years ago, considering the sediment layering on the ocean floor. Now the Younger Dryas dates about 12,700 - 11,650 - not nearly the same. I have said before, we have no idea what the Younger Dryas was in the first place, A heinrich event seems rather unlikely.

    About the Eart magnitic field, I surprised that Gary Glatzmaier did not show up yet in the thread. He managed to model geomagnetic collapses/reversals but still without the essential effects of the Earth precession and obliquity cycle

  16. Mar 10, 2007 #15
    Alternative Source for the Geomagnetic field?

    Hi Andre,

    Why do we believe the earth's core is the source for the geomagnetic field?
    Is there another conductive fluid on the earth that moves? Could electric currents in it, and motion of that conductive fluid create the geomagnetic field?

    Tinsley and Yu's paper (see attached link) discusses a "Global Electric Circuit" and a "Global Electric Current" that is driven by the solar wind (solar wind creates potential difference from Pole to Equator) and by thermal dynamic motion of the atmosphere (creates charge difference from cloud top to earth's surface). See page 4, figure 3.1. Do you see what I am driving at?

    Electric currrents in that fluid could create a magnetic field that would by induction create electric currents in the conductive liquid core. With that mechanism, energy is transferred from the earth's surface to the core, which would explain the thermal dynamic core anomalies that is a hot topic for geophysicists. (Cannot explain why the core is not cooler, based on current estimated temperature, thermal evolution of the core, measured heat loss of core to surface, and so forth.)

    http://www.utdallas.edu/physics/pdf/Atmos_060302.pdf [Broken]

    Before answering yes or no to the possibility of an alternative hypothesis, consider other data. How has the geomagnetic field varied with time? Attached is Valet's review paper, "Time Variations in the Geomagnetic Field Intensity".


    See page 22, Figure 9 b), Geomagnetic field variance over the last 75 kyrs. The graph compares geomagnetic field intensity determined from volcanic flows and sea floor sediment.

    41 kyrs ago the geomagnetic field intensity, has 75% less (2 x 10^22 A-m^2) as compared to the current geomagnetic field intensity of 8 x 10^22 A-m^2. Notice the geomagnetic intensity gradually increases to a peak of 12 x 10^22 A-m^2. This data is consistent with Yamazaki’s paper “Orbital Influence on Earth’s Magnetic Field 100kyr Periodicity”, see attached abstract. Paper finds 100 kyr and 41 kyr periodicity in the geomagnetic field. Yamazaki and others have found the geomagnetic field intensity varies with the inclination of the earth. What is causing the periodic geomagnetic field variance?

    Last edited by a moderator: May 2, 2017
  17. Mar 11, 2007 #16
    There's no doubt that the solar flare activity plays a role in Earth's total field, it is responsible for the external field component that geophysicists call diurnal variation. It does have a bearing on the electrical potential state of the earth right through to the core which effects the main field, this has been measured at different periods and we call this the magnetic response function, we can use it to infer electrical properties of the deep earth.

    Problems I foresee about saying this is the cause of earth's magnetism include: the atmosphere is not at all conductive therefore it is an unsuitable candidate as your fluid, the ocean on the other hand is conductive, but then how do you explain the dipolar nature of the field?

    Your problems about heat in the inner core can be solved if you allow the inner core to contain some K, this decays and heats things up, however the main source of the heat is the latent heat released by the crystallisation of the inner core, the expulsion of light elements at the ICB also drives compositional convection currents. Interesting fact: did you know that the viscosity of the outer core is about the same as water?
  18. Mar 11, 2007 #17
    The Limitations of Geodynamo Computer Models

    In reply to Andre's comment concerning a computer model that produces results that match some aspects of the real geomagnetic field.

    1) Computer models and the dynamo mechanism hypothesized cannot model the full period, during which the earth's magnetic field was observed. ie. There was been a geomagnetic field for 3.2 billions years based on the geomagnetic record. It can be surmised that the geomagnetic field existed earlier, say as early as 4 billion years ago, as it is believed the solar wind will strip off the earth's atmosphere without the geomagnetic field.

    Justification for Comment 1.
    The geodynamo models do not work before the core crystallized. The start of core crystallization has been set based on theoretical thermal dynamic calculations at a maximum 1.5 billion years ago to as early as 500 million years ago, even with radioactive Potassium to generate heat in the core to prevent the core from crystallizing earlier. ie. If the core crystallized earlier it would be larger than it is now. The crystallization creates a sharp localized temperature step which is required to produce convection currents to drive the dynamo.

    Comment 2: The current geodynamo models do not use real parameters, due to computer limations.

  19. Mar 13, 2007 #18
    What you say about models is true, we can't use grids to simulate outer core convection.

    I was wondering what you thought of the observed correlation between the magnetic field reversal paths and anomalously fast seismic regions in D''?
  20. Mar 16, 2007 #19
    Alternative Hypothesis for Source of Geomagnetic Field

    For both the "Core Only" geomagnetic field hypothesis and the "Core forced by Ocean" geomagnetic hypothesis, "the resulting large scale magnetic field is eventually produced and maintained by the mechanism of inverse cascade, whereby the field growth at small scales is arrested through the back action of the Lorentz force on the flow, and magnetic energy is continuously transferred towards larger and larger scales, eventually concentrating at the largest scale allowed by the geometry of the flow" (Ryskin's comment, from source Frisch et al, Fluid Mech. 68, 769 (1975) & Firsch and Leorat, Fluid Mech. 77,321 (1976).

    The above mechanism is hypthosized to be the same mechanism that creates the sun's and the galaxy's large scale magnetic field.

    The problem, in the case of the earth is the energy of the geomagnetic field is 0.8 x 10^18 Joules which is roughly a 1000 times less than the estimated energy in the fluid-core (Ryskin's comment, from source Roberts and Glatzmaier, Rev. Mod. Phys. 72, 1081 (2000). Theoretically it is expected that the energy of driver and field are roughly equal, which is the case for the galaxy and other examples.

    The energy of the ocean from wind, temperature, and tidal driven motion is at least estimated to be equal to the energy of the geomagnetic field. (Ryskin's estimate does not include the energy in the global electric current which provides the current source for the ocean mechanism.)

    The facts seem to support the assertion that the core only hypothesis has fundamental problems and that the alternative ocean forced core hypothesis is possible. Let's use Faraday's method, were we assume the alternative hypothesis is correct and then analyze the problem holistically to attempt to explain all past behaviour.

    Phenomena to explain:

    1. Very rapid and large geomagnetic field changes.
    2. A reduction of the geomagnetic (75% of current level), followed by a slow increase to current level.
    3. Magnesisferrite spinels (approx. 10 micrometers size) at minimums of the geomagnetic field. "The presence of ... magnesisferrite is ... a marker for ... a minimum in the intensity of the Earth's magnetic field". Preisinger et al. GSA Spec. Paper 356, page 213 (2002).
    4. Superchron periods: Multi-million year periods when there is no reversal of the magnetic field.
    5. Correlation of the planetary temperature and magnetic field level. (Which is affecting,
  21. Mar 22, 2007 #20
    I c..so if these two are different, y the emphasis is only on how magnetic fields are produced and not how gravitational fields are produced?? Also is there any way to create a gravitational field in a lab the way we create a magnetic field...?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook