Neutrino tomography of Earth


149 points | by lainon 1003 days ago


  • jboggan 1002 days ago
    This is really brilliant, literally jaw-dropping. I was really blown away by the idea and construction of the detector itself and this application is astounding.

    • bacon_waffle 1002 days ago
      My favourite story about the construction of IceCube relates to the drilling of the holes to deploy the detector. They used two types of drill, which melt the firn (compacted snow) on top and ice lower down. One is a sort of conical heat exchanger, the other circulates water through a Big water heater on the surface.

      In the early days of South Pole Station (late 1950s), supplies were airdropped in - the cargo planes of that era couldn't land on the snow and going overland was very difficult. Many of the parachutes wouldn't open properly, and so lots of supplies (including a bulldozer) and parachutes wound up buried in the snow. South Pole slowly accumulates snow, so that debris has gotten deeper and deeper below the surface over time.

      Of course, the holes required to build IceCube had to go deeper than those supplies, but the drill can't go through things that don't melt, so in some cases the holes had to be moved off the nominal grid. According to the story I heard, one of the holes ran in to a bunch of meat, which floated to the top of the bore...

    • gpuhacker 1002 days ago
      You may like KM3NeT then ( The goal is to build a cubic kilometer spanning instrument at the bottom of the Mediterranean.
    • ArtWomb 1002 days ago
      Agreed. This is truly astounding. Neutrinos seem to encapsulate an unnaturally high degree of complexity and information in such a tiny package ;)

      Formalism of Neutrino Oscillations

      • oh_sigh 1002 days ago
        I always thought that if an alien civilization is trying to communicate with earth, they would do it via neutrinos and not radio waves, because if you know where you are communicating to, you don't need to worry about signal attenuation as much as with EM waves.
    • Balgair 1002 days ago
      Supplementary Figure 6 is pretty cool! There seems to be a lot of strange things happening at the mantle-crust boundary that will take some further study to understand.

      All said, this is a MAJOR advance in our understanding of the Earth and the science thereof. This technique should allow for some pretty amazing things in the future.

      • phkahler 1002 days ago
        Would it be feasible to put a source and detector in orbit so we could get high resolution CT images of the interior of the earth? IIRC the detector used here is very large, but there are smaller ones and a directed source would go a long way towards making the detection simpler.
        • InclinedPlane 1002 days ago
          Neutrino detection generally requires large volumes and large masses (for very large values of "large"), neither of which are as yet feasible for space based systems. The IceCube neutrino observatory, for example, makes use of a volume of ice that is on the order of a cubic kilometer (over 900 million tonnes). Suffice it to say, building cubic kilometer solid structures in space that weigh hundreds of millions of tonnes is outside of our engineering capabilities at present. However, within the next several decades this might be feasible.
          • godelski 1002 days ago
            That's also understating the size. While the final detection is happening in the ice, the column of air above it is also important.
            • lootsauce 1002 days ago
              Could we not just land a detector on the far side of the moon. Thats pretty large no?
              • pvg 1002 days ago
                It is but you need both a big hunk of mass and a way to detect the neutrino interactions inside it. Ice Cube is shot through with detectors. One of the OG detectors measured the particular isotope created during interactions.


                The Super-Kamiokande has photomultipliers staring at a bunch of water looking for Cherenkov radiation resulting from neutrino interactions


                None of these are much of a practical option with the moon.

                • InclinedPlane 1002 days ago
                  You need to have the reaction mass to make use of. Most neutrino detectors use special chemicals in custom tanks, IceCube uses antarctic ice. You would need not just mass but ice or water for a neutrino detector, ideally highly compressed optically clear ice. The only possible candidate in the Solar System that fits that bill at present is Europa. Though it's possible that similar conditions might also exist on Mars (we know there are sub-surface glaciers, but we don't know how deep they go or how pure the ice is. And potentially some of the large sub-surface oceans (on Ganymede, Enceladus, etc.) might work too but we know even less about their properties.
                  • pvg 1002 days ago
                    Not all detectors were optical so if you found a moon made of, say, dry cleaning fluid, you'd still be good.
                  • phinnaeus 1002 days ago
                    Each detector has a limited range. And only works well in certain mediums, generally water or water ice. Even if the Moon was solid ice, we would need to surround it with detectors, as well as drill a huge number of holes all the way through it and place detectors (probably hundreds of thousands or millions) in each hole.
                • godelski 1002 days ago
                  I had this idea back in my undergrad when my professor first told me about neutrinos (which was his forte). The harder I looked the more infeasible it seemed. The big problem, is creating a detector. The reason this is a big problem is because neutrinos are so small and so neutral. How small? An electron is 511keV. A neutrino is <= 0.21eV. That's 7 orders of magnitude! And on top of that, they have a neutral charge. There is no electric dipole moment either. So how does it interact? Basically by striking another particle. And remember that atoms are basically empty space. So chances of that are slim.

                  So to place a detector in orbit we'd need one of two things. Space travel to become cheap enough where putting a massive body up is cost feasible, or a radically different understanding of how to detect neutrinos. (First is more likely)

                  The former also comes with some problems. You'll find more about this by asking why detectors are so far underground. And there's a big connection there to generating the resolution you'd need to map the interior of the Earth.

                  • gpuhacker 1002 days ago
                    An in space detector is (at least to my understanding) not that easy, as you need the surface of the Earth to interact with the neutrinos. These interactions result in muons, which allow to detect that there was a neutrino in the first place. Also, there is a pretty big source for neutrinos pretty close to Earth, but Earth is in its orbit, as opposed to the other way around.
                    • not2b 1002 days ago
                      The IceCube neutrino detector consists of about a cubic kilometer of ice. You're not going to put one of those in orbit.
                      • njarboe 1002 days ago
                        But it would make a cool joint project for NASA and the NSF to fund. Capture a comet big enough that you could mine enough water to make a cubic kilometer sized detector in Earth orbit.
                        • BurningFrog 1002 days ago
                          That could probably also wipe out half a continent if you mis-aim it...

                          Then again, it would look spectacular in the sky!

                        • godelski 1002 days ago
                          There are other types of neutrino detectors. Not that it really makes the problem any easier to solve, but this isn't the only way.
                          • phkahler 1001 days ago
                            That's what I was thinking. They have sent neutrinos from one lab to another with a man-made source and a detector smaller than 1km^3. The idea is to put a strong source in orbit and aim it at an orbiting detector. Over time this configuration would be used to scan the entire planet. Still a huge challenge.
                            • godelski 1001 days ago
                              That's essentially the main idea most people come to. Using a source so you can concentrate on specific ones and not get distracted by solar, cosmic, or terrestrial neutrinos. Or you could just use terrestrial ones. But still, that detector is the hardest part to figure out.
                          • gpuhacker 1002 days ago
                            • logfromblammo 1002 days ago
                              You might be able to tow it in from elsewhere using a light sail, however. Beyond our current technology, but might be doable by 2100, or sooner with a fortuitous comet trajectory.
                              • dotancohen 1001 days ago
                                And once it is in orbit, how do you propose stationkeeping against exospheric drag? Preferably a solution that will remain reliable for 10^n (n>>3) years.
                                • logfromblammo 1001 days ago
                                  A cubic kilometer of ice is 9.167e11 kg.

                                  Also, just keep the orbit above 800km, and the light sail will still be able to counteract its own drag.

                                  If the orbit gets too low, turn the ice into rocket fuel and use it to refuel ships boosting into higher orbits.

                                  • dotancohen 1001 days ago
                                    I'm pretty sure that a cubic kilometer of ice in orbit would have more value in other purposes than as a neutrino detector!
                                    • logfromblammo 1000 days ago
                                      Sure, but as long as it's up there, might as well use it in every way possible.
                          • cozzyd 1002 days ago
                            The muons are produced in the upper atmosphere (from cosmic rays interacting with the atmosphere). You can't do this in space. Not too mention the size of a neutrino detector, although you can use the e.g. regolith of the moon as a neutrino detector (via the Askaryan effect). In fact there are experiments pointing radio telescopes at the moon to look for ultra-high-energy neutrino interactions. My not-so-polite term for them is lunatics :).
                      • lolc 1001 days ago
                        Nice but I'm still left wondering what the J/bit of neutrino transmission would be. Wouldn't it be cool if we could replace the long fiber optics stretching around the globe with direct-through neutrino stations?
                        • russdill 1001 days ago
                          Given the difficulty in neutrino detection, I seriously doubt it would replace fat links. You could gain quite a bit of advantage when it comes to trading though using a low bandwidth link.
                      • peter303 1002 days ago
                        Frankly this idea has been repeated every few years since the invention of tomography 40 years ago. Hopefully someone will actually do it one of these years.
                        • caf 1002 days ago
                          In theory, would IceCube be able to detect neutrinos generated by the same mechanism in the atmosphere of Venus?
                        • k_sze 1002 days ago
                          Could this technique be useful on other planets?
                          • stocknosticator 1002 days ago
                            • richardbdrowley 1002 days ago
                              > Neutrino detection will revolutionize oil and other natural resource discovery.


                              • godelski 1002 days ago
                                To be fair, it probably will. But the ability to do it in a cheap enough manner is certainly far enough away that the value of those resources will likely be much lower than they are today.

                                Accurate, cheap, and small neutrino detectors would revolutionize any industry dependent upon locating things within... well... anything. But that statement is as meaningful (if not less) as "faster computers will revolutionize the automotive industry."