Taking a Picture of the Earth's Interior with Geoneutrinos

The Earth contains a certain amount of natural radioactivity, and the decay of these radioactivities is an important and perhaps dominant source of geothermal heat. These same decays also generate particles known as neutrinos. In measuring the arrival directions of neutrinos generated in the decay of natural radioactive elements in the Earth's interior, it will be possible to get a three-dimensional picture of the Earth's composition and shell structure. This will provide a new and detailed understanding of the origin of the Earth's geothermal heat, and will finally answer the question of how much heat comes from radioactive decays, and how much is "primordial" heat leftover from the birth of the Earth. The mapping of the Earth's interior might also help give answers to such questions as "What powers the magnetic field of the Earth?" and "What dominates the geodynamo?". To actually take a neutrino picture of the Earth is quite challenging technically, but not impossible, and we are optimistic about prospects for taking the first images in the near future.

Neutrinos are subatomic particles created in certain types of nuclear decays. These ghostly particles have no electric charge, a tiny but nonzero mass, and are very weakly interacting. A typical neutrino passes through matter unseen, unfelt, and unperturbed. Indeed, each second of the day and night, many billions of neutrinos pass through you harmlessly; these neutrinos originate mostly from the Sun, but also from the Earth's atmosphere, and from the interior of the Earth. These last "geoneutrinos" have received a great deal of attention lately, and open up new views onto the Earth.

Geoneutrinos are produced in the decays of unstable, radioactive elements--mostly uranium, thorium, and potassium (40K)--inside the Earth. These same decays also generate heat, which makes up some portion (thought to be about 60%) of the geothermal heat flow. The amount of geoneutrinos (and radioactive geothermal heat) depends on the amount of radioactive material through the Earth. We do know very well the amount and distribution of radioactive material in the Earth's thin crust. However, we have only "scratched the surface" of the Earth with direct measurements: the deepest probes that have been recovered from drill holes in the earth's crust come from a depth of only about 10 km, and volcanos bring up mantle material from a depth of a few hundred kilometers. So most of the earth's interior is hidden! Consequently, it remains unclear how much radioactive material is in the inside the bulk of the earth and where it is located. In particular, it is unclear how abundant radioactive elements are in the very core of the Earth. The standard geophysical prediction is that the core is void of all radioactivity. Recently, however, there have been some suggestions that there may be a significant amount of potassium in the Earth's core. These uncertainties about the Earth's radioactive content translate into uncertainties about the amount of geothermal heat and geoneutrinos that are generated by radioactive decays. Thus, measurements of geoneutrinos can address these questions and ultimately determine the nature of the radioactive earth.

Recently the Kamland experiment, which was primarily designed to measure anti-neutrinos from nuclear reactors, reported 9 events due to geoneutrinos. This marks the first detection of neutrinos from the Earth's interior, and already demonstrates that radioactivity is an important heat source for the Earth. This promising discovery gave rise to several papers about the possibilities a measurement of geoneutrinos could provide.

In a 2003 paper, the Italian collaboration of Mantovani, Carmignani, Fiorentini and Lissia created a model for the Earth's interior and calculated from it the expected number of geoneutrinos passing through the detector (i.e., the neutrino flux). This model is based on present geological data of the density structure and radioactive element abundance and distribution in the Earth's crust and mantle. It is presently controversial whether the Earth's core contains radioactive material. While the standard picture calls for none, several recent laboratory experiments suggest the possibility that a large amount of potassium could have formed alloys with iron when the earth was created.

stack In our new paper, "Imaging the Earth's Interior: The Angular Distribution of Terrestrial Neutrinos," we calculated the pattern of arrival directions (i.e., the angular distribution) of the geoneutrinos. Simply put, we simulated the image of the Earth as seen by an observer with "neutrino eyes." To so this, we used geophysical models of the Earth's interior (the Mantovani, Carmignani, Fiorentini & Lissia model for the crust-mantle system and the predictions of the laboratory experiments for the core) to create several models of the Earth's interior and its radioactive content. In all models, radioactive elements are highly concentrated in the outer crust of the Earth; the differences come in whether or not the core contains any radioactivity.

The images at left show two of our simulations of the neutrino views of the Earth. Imagine donning "neutrino goggles" and looking down; these images show what one would see. However, the Earth takes up half of the "sky" (i.e., the ground) and thus occupies a full hemisphere of your field of view. To represent a curved hemisphere on a flat image requires some projection effects (similar to those needed in representing the whole globe of the Earth on a flat map). In these images, the center is what you would see looking straight down--these are the neutrinos which come up through your feet! The outer edge of the disk represents the view towards the horizon, so that as you scan from the center to the edge of the image, imagine going from a downwards to a horizontal view. We see that in both cases, there is a large signal right near the horizon--this is due to the high concentration of radioactive species in the Earth's crust. These neutrinos do not come through your feet, but rather arrive almost horizontally--hitting you in the forehead! The main difference between the image the brightness of the center, which differs depending on the geophysical model we use to describe the Earth's interior.

The two pictures of the Earth show the extreme cases for the nature of the Earth's core. In the upper picture the center (core) is almost dark, as no radioactive material was included. The bottom picture has a bright center. Here the maximum amount of potassium given in the laboratory experiments (7000ppm) was added to the upper picture. Notice the large contribution of the crust in both cases! We find that even with a crude angular resolution of only 30 degrees a neutrino detector is able to distinguish between the models and to determine how much potassium is in the core. That is, even a very blurry neutrino view of the Earth can answer the question "is there a bright spot underfoot?".

Furthermore, we showed such a "neutrino picture" can be mathematically transformed to reveal the full three-dimensional distribution of radioactive elements inside the Earth. The procedure, a sort of "tomography" of the Earth, is mathematically analogous to that used in medical CAT scans to obtain three-dimensional images of the interior of the body. Thus, with a neutrino image of the Earth, we can map the Earth's radioactive content and structure. This will provide unique new information which will resolve the question of how much geothermal heat is actually generated by the Earth, and will also give unprecedented insight into the Earth's structure.

sun Can we actually take a real neutrino picture of the Earth? The task is difficult but not impossible. A right is a neutrino image of the Sun, taking by the Super-Kamiokande experiment, which detects regular (i.e., not anti-) neutrinos. This picture has an angular resolution of about 2 degrees, which would be fantastic for geoneutrinos (but is still larger than the Sun's disk as seen from the Earth). For antineutrinos, angular resolution is more challenging, but can be achieved.

The neutrino detector that attempts to measure the angular distribution has to be sensitive to low energies. A possibility would be to use a proton-rich scintillator enriched with gadolinium as detector material. An incoming neutrino reacts with a proton, which is the first light pulse. In this process a neutron is created, which will be absorbed by a gadolinium nucleus and therefore produces the second observable pulse. In these location of these two pulses directional information is hidden, as one can assume that the produced neutron moves in the same direction as the incoming neutrino. Although a lot of the directional information is lost in the scattering of the neutron before it is absorbed, an angular resolution of 20 degrees was observed in the CHOOZ experiment. This experiment is not running any more and was too small to be able to detect a noticeable amount of geoneutrinos. When it was running, CHOOZ was sensitive to the direction of the antineutrinos it received from a nearby nuclear reactor. CHOOZ was able to use the neutrino signal to recover the reactor direction to within 20 degrees. While this is quite "blurry" vision, such a "coke-bottle-glasses" view of the Earth would still be sufficient to distinguish between the two cases shown in the terrestrial images above.

Thus, the prospects for the future are exiting and promising: The SNO Collaboration has been thinking of using a scintillator with gadolinium. If they or other experiments are able to detect the angular distribution of geoneutrinos then the mystery of the interior of the earth, a planet so close and yet often so far beyond our knowledge, might finally be resolved.

Further Information

Physics

Geology

  • V. Rama Murthy, W. van Westrenen, Y. Fei: Nature 423 (2003)
  • K.K.M. Lee, R. Jeanloz: Geophys. Res. Lett 30 (2003)
  • C.K. Gessmann, B.J. Wood: Earth Planet. Sci Lett, 200, p.63-78 (2002)

Links

Further Information

Questions? Please contact Brian Fields or Kathrin Hochmuth

back to top
this page by Kathrin A. Hochmuth and Brian D. Fields
Last modified: Tue Aug 31 17:24:18 CDT 2004