Scientific Progress Goes “Chirp”!

 
640px-Northern_leg_of_LIGO_interferometer_on_Hanford_Reservation
Northern leg (x-arm) of LIGO interferometer on Hanford Reservation. By Umptanum – wikipédia, CC BY-SA 3.0.

Ladies and Gentlemen, we have detected gravitational waves! We did it!” And with that, the Laser Interferometer Gravitational-Wave Observatory (LIGO) announced the discovery of a strange feature of General Relativity that’s been hiding since the 1960s.

Gravitational waves are ripples in the fabric of space, caused by the acceleration of mass. The bigger the mass and the faster it’s accelerated, the better, and there’s nothing that better fits that description than two black holes orbiting each other. These waves expand and compress space, which slightly changes distances between objects in their path. Two objects floating freely in deep space far from any forces, would find themselves carried a little towards each other and then a little away from each other and back and forth, as the wave passed over them. If you could measure the distance between them accurately, like with a laser range-finder, you’d detect that oscillation. But this is tough to do on the Earth, because you can’t have the two objects floating freely any more. The best you can do is to hang them from fine threads, like a swing. They’re not free to move in every direction, but they can swing back and forth.

So, how did we find them? LIGO shines laser beams down two 4-kilometer (2.5-mile) vacuum pipes, arranged in an “L” shape. At the end of each pipe is a mirror that reflects the beam back and measures its length, with an accuracy of 10^-18 m, a thousandth the width of a proton. This always amazes me, because the mirror itself is made out of atoms, which are bumpy and much larger than protons. And yet, it works because the bumps average themselves out. When a gravitational wave comes by, it shrinks the distance along one pipe while expanding the distance along the other (as they’re at right angles) and then it switches.

What they’ve announced today is the detection of gravitational waves coming from the merger of two black holes, spiraling into each other a billion light-years away from us. At 29 and 36 times the mass of the Sun, these aren’t very big for black holes: the ones in the centers of galaxies have millions or billions of solar masses. But the smaller ones are more plentiful, which means they’re more likely to collide. As the two in question orbited each other, the gravitational waves carried away more and more of their energy, making them spiral inward, until they finally collided and released massive amounts of energy (the equivalent of the mass of three Suns converted directly into energy).

Take a look at the wave patterns that LIGO felt:Gravitational waves felt by LIGO

The vibration is vertical, and time goes to the right. Early on, the waves are slow (i.e., broad) and weak (i.e., short). As the black holes spiral inward, the waves strengthen and increase in frequency until they make a loud “chirp! when they collide. Then, there’s a faint wiggling at the end, as the new, combined black hole rings like a bell until it settles down.

This pattern is so clear, you can see it without needing the computer to sort it through for you. What are the odds of detecting this? My wife — who works in numerical relativity (the computer simulation of the waves) — tells me that it was entirely possible they’d run this for decades before they heard anything. Instead, they found this within days of turning the machine on. That suggests that these black holes are colliding all over the place, and we’re going to find plenty more.

So what good is this? The press is mostly writing headlines shouting, “Einstein is proven right!” This is true in a simplified sense, but not too many of us were expecting otherwise and that’s not really why all of this was built. With more detections and better signals, we’ll get to the point where we can test theories that go beyond General Relativity, which is interesting. But I think the real value of this discovery will be observational.

Until now, astronomy has almost exclusively relied on what we can see with light: i.e, electromagnetic radiation, ranging from radio waves on one end to gamma rays on the other. We can see the gas falling into black holes — glowing x-ray hot before it disappears — and we can see supernovas from the light they emit. But there’s a lot going on down in the compact centers of these things that we can’t see with light. Gravitational waves will give us an entirely different way to probe them.

Already, the gravitational wave people and the visible light astronomers have teamed up to do joint searches for flashes of light that might accompany a gravitational wave burst. My wife, for example, is working on the computer simulations for these. This photo shows her working out the algorithm for her models.

Maria Babiuc Hamilton, working out her numerical relativity algorithm.

The discovery of gravitational waves has been a long time coming. Searches started in the 1960s, and there’s even a detector on the Moon, placed there by Apollo 17. But the evidence up to this point had always been indirect. Now, we’ve finally got something both direct and obvious.

That’s big news.

Members have made 108 comments.

  1. 1
  2. 2
  3. 3
  4. 4
  1. Profile photo of PHCheese Member

    It is of my opinion that a phenomenon such as this could be as much responsible for changes in the earths temperature over time as anything else.

    • #1
    • February 11, 2016 at 4:46 pm
    • LikeLike
  2. Profile photo of Tim H. Member
    Tim H. Post author

    While I was writing this up, I felt some literary inspiration and came up with this. Just for fun.

    • #2
    • February 11, 2016 at 4:53 pm
    • LikeLike
  3. Profile photo of John Walker Contributor

    Tim H.: But I think the real value is observational. Up ’til now, astronomy has almost exclusively relied on what we can see with light: electromagnetic “radiation.” We can see the gas falling into black holes, glowing x-ray hot before it disappears, and we can see supernovas from the light they emit.

    I’ve been calling this the “third channel”. The first channel is electromagnetic radiation, which we’ve been observing with our eyes since antiquity and more recently with instruments in other parts of the spectrum. The second channel is particles, which as cosmic rays, we’ve been observing since the early 20th century. Today ultra high energy cosmic ray observatories and neutrino detectors provide another window into the universe. Today the opening of the third channel was revealed. Gravitational waves provide access to phenomena inaccessible to electromagnetic radiation and particles because the universe is essentially transparent to gravity (which is why gravitational waves are so difficult to detect).

    Today’s publication was of an event precisely as expected from astrophysics and general relativity. There is much more excitement to come when, as with the opening of the first two channels, we begin to observe things we never anticipated seeing.

    • #3
    • February 11, 2016 at 5:01 pm
    • LikeLike
  4. Profile photo of BrentB67 Inactive

    This is great stuff, I think.

    Now for the smart aleck question of the day.

    We can have lasers in L-shaped vacuum tubes shone on mirrors measuring distance a thousandth of the width of a proton and prove gravitational waves, but we have to work the math out on a blackboard?

    Are there no whiteboards? Just sayin’.

    • #4
    • February 11, 2016 at 5:12 pm
    • LikeLike
  5. Profile photo of Hank Rhody Member

    BrentB67:This is great stuff, I think.

    Now for the smart aleck question of the day.

    We can have lasers in L-shaped vacuum tubes shone on mirrors measuring distance a thousandth of the width of a proton and prove gravitational waves, but we have to work the math out on a blackboard?

    Are there no whiteboards? Just sayin’.

    Now for the smart aleck answer:

    Racist.

    • #5
    • February 11, 2016 at 5:23 pm
    • LikeLike
  6. Profile photo of Larry Koler Member

    John Walker: Today’s publication was of an event precisely as expected from astrophysics and general relativity.

    Isn’t that convenient for them. What are the chances?

    Confirmation bias is hard to watch, isnt’t it? Oh yes, but Einstein, Larry. And math, Larry.

    • #6
    • February 11, 2016 at 5:25 pm
    • LikeLike
  7. Profile photo of Hank Rhody Member

    Tim H.: My wife tells me that it was entirely possible they’d run this for forty years before they heard anything at all. Instead, they found this within days of turning the machine on.

    A decade ago when I was studying physics at the University of Wisconsin Milwaukee I heard a lot about this thing. We were an associated school. I remember seeing a presentation where they had a lovely blank graph of the results so far, but not to worry! here’s what the simulation produced. I remember a grad student chortling about how the theorists had proposed the thing, and built it, without taking into account the thermal properties of the string you had to use to suspend the weights.

    I said they talked about it a bunch, I didn’t say I listened.

    • #7
    • February 11, 2016 at 5:34 pm
    • LikeLike
  8. Profile photo of John Walker Contributor

    Larry Koler:

    John Walker: Today’s publication was of an event precisely as expected from astrophysics and general relativity.

    Isn’t that convenient for them. What are the chances?

    Confirmation bias is hard to watch, isnt’t it? Oh yes, but Einstein, Larry. And math, Larry.

    Please read the paper. The event popped out of a transient filter which simply measured events larger than the noise threshold of the instrument with no assumptions about the signal. It was only then matched to a filter based upon the models of general relativity and found to closely conform to the inspiral of two black holes, confirmed by the profile of increasing frequency and intensity of the signal, followed by its abrupt end, which the model interprets as the coalescence of the two black holes into one larger Kerr (spinning) black hole.

    This result surprised the investigators. They assumed the first events they’d see would be inspirals of two neutron stars, or maybe a neutron star into a black hole, since such systems were believed to be much more common that black hole binaries. Further, before this event, there was no direct evidence for intermediate black holes with the mass of those observed in this event.

    The fact that the same waveform was observed in Hanford, Washington and Livingston, Louisiana, with precisely the speed of light offset expected for propagation of the wave between the two detectors rules out almost all hypotheses other than a “conspiracy so vast…”.

    In the next five years, we’ll observe dozens of these events, begin to characterise them, and test them against models of their causes.

    • #8
    • February 11, 2016 at 5:39 pm
    • LikeLike
  9. Profile photo of LC Member
    LC

    That chirp was amazing to hear this morning in the press conference. On top of the already great plots.

    • #9
    • February 11, 2016 at 5:42 pm
    • LikeLike
  10. Profile photo of John Walker Contributor

    Hank Rhody: I remember a grad student chortling about how the theorists had proposed the thing, and built it, without taking into account the thermal properties of the string you had to use to suspend the weights.

    Eric Drexler observed, “Scientists do engineering in order to do science, while engineers do science to do engineering.”

    • #10
    • February 11, 2016 at 5:45 pm
    • LikeLike
  11. Profile photo of Carey J. Inactive

    Do we have any clue as to the propagational speed of gravitational waves? Is it equal to the speed of light? Is it slower? Is it faster? If it is different, and you have a way to identify an EM-observable event that created them, you can determine the distance to the event. Kind of like timing the lag between a lightening strike and its thunder.

    • #11
    • February 11, 2016 at 5:48 pm
    • LikeLike
  12. Profile photo of John Walker Contributor

    Carey J.:Do we have any clue as to the propagational speed of gravitational waves? Is it equal to the speed of light? Is it slower? Is it faster? If it is different, and you have a way to identify an EM-observable event that created them, you can determine the distance to the event. Kind of like timing the lag between a lightening strike and its thunder.

    According to general relativity, the propagation of gravitational waves is the same as the speed of light. This is a fundamental part of the theory (powers of the speed of light appear all over the equations), so any deviation from this would falsify the theory.

    This event allows constraining the mass of the graviton (the particle by which, in quantum mechanics, gravity is propagated) to a very low upper limit (1.2×10^−22 eV/c²). This means that, since in relativity, a massive particle must always propagate slower than light, any deviation between the speed of propagation of gravity and light must be very small. There are abundant theoretical reasons to believe that gravity and light propagate at the same speed, but here I’m talking solely about experimental measurements.

    • #12
    • February 11, 2016 at 6:11 pm
    • LikeLike
  13. Profile photo of Midget Faded Rattlesnake Moderator

    BrentB67:This is great stuff, I think.

    Now for the smart aleck question of the day.

    We can have lasers in L-shaped vacuum tubes shone on mirrors measuring distance a thousandth of the width of a proton and prove gravitational waves, but we have to work the math out on a blackboard?

    Are there no whiteboards? Just sayin’.

    What I was told is that whiteboards are allegedly better for any computers occupying the room.

    But there’s something about a blackboard… maybe it’s a tactile thing…

    • #13
    • February 11, 2016 at 6:49 pm
    • LikeLike
  14. Profile photo of BrentB67 Inactive

    Midget Faded Rattlesnake:

    BrentB67:This is great stuff, I think.

    Now for the smart aleck question of the day.

    We can have lasers in L-shaped vacuum tubes shone on mirrors measuring distance a thousandth of the width of a proton and prove gravitational waves, but we have to work the math out on a blackboard?

    Are there no whiteboards? Just sayin’.

    What I was told is that whiteboards are allegedly better for any computers occupying the room.

    But there’s something about a blackboard… maybe it’s a tactile thing…

    When I saw the picture I was trying to think of the last time I used a blackboard. I would think the dust in laboratory setting would be bad, but maybe we have dustless chalk.

    • #14
    • February 11, 2016 at 7:05 pm
    • LikeLike
  15. Profile photo of BrentB67 Inactive

    As these waves propagate through space would it be possible to ‘catch’ one, ride it, or in any way augment space travel?

    • #15
    • February 11, 2016 at 7:06 pm
    • LikeLike
  16. Profile photo of Tom Meyer, Common Citizen Contributor

    Many, many thanks to Tim for writing this.

    Tim H.: As the two in question orbited each other, the gravitational waves carried away more and more of their energy, making them spiral inward, until they finally collided and released massive amounts of energy (the equivalent of the mass of three Suns converted directly into energy).

    I can’t really wrap my head around how little I can wrap my head around that.

    That’s just an insane amount of energy.

    • #16
    • February 11, 2016 at 8:00 pm
    • LikeLike
  17. Profile photo of Mark Wilson Member

    Tim H.: What they’ve announced today is the detection of gravitational waves coming from the merger of two black holes, spiraling into each other a billion light-years away from us.

    I also do modeling, prediction, and data matching in my job. I’m curious about the signals called “Prediction”. Is that a prediction based on two known black holes we knew were going to collide, then measured and saw the data matched? Or as often happen, did they see some data and then try to create a post-event “prediction” by finding the right parameters to populate the model and get it to match the data?

    • #17
    • February 11, 2016 at 8:21 pm
    • LikeLike
  18. Profile photo of Mark Wilson Member

    John Walker: The fact that the same waveform was observed in Hanford, Washington and Livingston, Louisiana, with precisely the speed of light offset expected for propagation of the wave between the two detectors

    I’ve always been curious about the distant clocks used in large scale physics experiments. I assume they use GPS as a time source, but what do they do about the vastly different iono- and tropo- delays with different climates, different solar exposure, and different constellation visibility to get such a precise degree of synchrony?

    • #18
    • February 11, 2016 at 8:33 pm
    • LikeLike
  19. Profile photo of Tom Meyer, Common Citizen Contributor

    At a friend’s suggestion, we once drove along WA-240, which skirts Hanford. It’s the kind of place you expect to run into Mulder and Scully.

    • #19
    • February 11, 2016 at 8:40 pm
    • LikeLike
  20. Profile photo of drlorentz Member

    Hank Rhody: I remember a grad student chortling about how the theorists had proposed the thing, and built it, without taking into account the thermal properties of the string you had to use to suspend the weights.

    To be fair to these guys, the effect they are seeking is so small that they need to worry about all kinds of issues that could normally be neglected. For example, the test masses (mirrors in the case of LIGO) have to be extremely uniform and pure because otherwise they will have greater thermal motions that mask the signal. A former colleague did her Ph.D. thesis on minimizing the losses at the attachment points to suspend the masses.

    No one person can be expert in all these areas, so it’s hardly surprising that the theorists you heard didn’t think of everything. These are not simply engineering problems. To even know what to look for you need to understand things like the fluctuation-dissipation theorem, which most engineers have never heard of.

    • #20
    • February 11, 2016 at 8:49 pm
    • LikeLike
  21. Profile photo of drlorentz Member

    Mark Wilson:

    I’ve always been curious about the distant clocks used in large scale physics experiments. I assume they use GPS as a time source, but what do they do about the vastly different iono- and tropo- delays with different climates, different solar exposure, and different constellation visibility to get such a precise degree of synchrony?

    The two LIGO detectors are separated by 10 milliseconds. GPS clock signals are accurate to tens of nanoseconds, about a factor of a million smaller. Differential time-of-flight delays through the ionosphere are on the order of nanoseconds.

    As an aside, GPS must account for General Relativity corrections to be accurate. This is weak-field relativity, which is considered to be well verified. The waves LIGO detected were probably the first strong-field test of relativity.

    Edit: The actual time difference between the two arrival times was about 7 ms according to the paper. The arrival times have to be within the 10 ms spacing of the two detectors but can be less depending on the angle of arrival of the wave. The paper says that the data acquisition systems are synchronized to within 10 μs using GPS.

    • #21
    • February 11, 2016 at 8:59 pm
    • LikeLike
  22. Profile photo of James Lileks Contributor

    Stupid question from a layman: “Smaller black holes are more plentiful” made me wonder if that might be where the missing mass is. I’m sure it’s occurred to them, but is there any solid refutation of the possibility?

    There’s a pretty good chance I won’t be around to witness the end of the cosmos by Heat Death, but I just hate that idea. Much prefer that there’s sufficient mass so everything comes back home, compacts, the explodes again, over and over, regular as respiration.

    I’m fine with the multiverse, though.

    • #22
    • February 11, 2016 at 9:34 pm
    • LikeLike
  23. Profile photo of Z in MT Member

    Larry, as I told a colleague this morning I don’t trust the astronomers. Most astronomers I know are happy to calculate something to a factor of pi. However, I do trust the scientists and engineers at LIGO. Many of them have a background in time and frequency metrology, and you will never meet a more cautious bunch of perfectionists.

    • #23
    • February 11, 2016 at 9:35 pm
    • LikeLike
  24. Profile photo of Randy Webster Member

    BrentB67:This is great stuff, I think.

    Now for the smart aleck question of the day.

    We can have lasers in L-shaped vacuum tubes shone on mirrors measuring distance a thousandth of the width of a proton and prove gravitational waves, but we have to work the math out on a blackboard?

    Are there no whiteboards? Just sayin’.

    And just look at that piece of chalk she’s using!

    • #24
    • February 11, 2016 at 9:36 pm
    • LikeLike
  25. Profile photo of drlorentz Member

    Z in MT:Larry, as I told a colleague this morning I don’t trust the astronomers. Most astronomers I know are happy to calculate something to a factor of pi. However, I do trust the scientists and engineers at LIGO. Many of them have a background in time and frequency metrology, and you will never meet a more cautious bunch of perfectionists.

    Years ago, the joke was that precision cosmology was an oxymoron. I had the same feeling about astronomy in general This changed when good measurements of the Hubble constant were made in the 1990s. They’re doing better now. Anyway, as you noted, many of the people are working on this are optical physicists anyway.

    • #25
    • February 11, 2016 at 10:50 pm
    • LikeLike
  26. Profile photo of Dustoff Inactive

    I’m pretty much with James Lileks on this one.

    • #26
    • February 11, 2016 at 11:08 pm
    • LikeLike
  27. Profile photo of Peter Meza Member

    OK, I get that Einstein predicted gravitational waves.

    I get that the LIGO detector captured an event.

    I also see a “predicted wave” superimposed on the discovered wave.

    In what sense where the waves we see in the graph “predicted”. How where they predicted? You mean this exact event had a predicted signature? As in this predicted wave signature was in the computer a couple of years ago, and now the signal was actually detected and jeez look at how similar they are?

    I am missing something.

    • #27
    • February 11, 2016 at 11:16 pm
    • LikeLike
  28. Profile photo of Mark Wilson Member

    Peter Meza: In what sense where the waves we see in the graph “predicted”. How where they predicted?

    My suspicion is that they retroactively used a simulation and fiddled with the inputs until the simulation could generate a “prediction” that matched the data. If so, it’s not a prediction in the traditional prescient sense, but in the model-agrees-with-reality sense.

    • #28
    • February 11, 2016 at 11:28 pm
    • LikeLike
  29. Profile photo of drlorentz Member

    Mark Wilson:

    Peter Meza: In what sense where the waves we see in the graph “predicted”. How where they predicted?

    My suspicion is that they retroactively used a simulation and fiddled with the inputs until the simulation could generate a “prediction” that matched the data. If so, it’s not a prediction in the traditional prescient sense, but in the model-agrees-with-reality sense.

    Yes, it’s a loose use of the term predicted, but not out of line with common usage in physics. It’s reasonable since the signature correspond to the kind of event they were looking for. There are some adjustable parameters but not too many.

    Put it this way, there’s a family of waveforms that can be simulated for events of this kind. They all look similar. You can’t predict the exact parameters for some signal that hasn’t arrived yet but you can make the prediction that it’s going to be a member of that family. I think that’s a reasonable use of the word predicted.

    Edit: The paper does not use the word predicted in the graphs. The figures in the OP must have been from somewhere else. The paper says,

    It matches the waveform predicted by general relativity for the inspiral and merger of a pair of black holes and the ringdown of the resulting single black hole.

    and

    to recover signals from the coalescence of compact objects, using optimal matched filtering with waveforms predicted by general relativity.

    figs

    • #29
    • February 11, 2016 at 11:53 pm
    • LikeLike
  30. Profile photo of Z in MT Member

    Peter Meza:I also see a “predicted wave” superimposed on the discovered wave.

    In what sense where the waves we see in the graph “predicted”. How where they predicted? You mean this exact event had a predicted signature? As in this predicted wave signature was in the computer a couple of years ago, and now the signal was actually detected and jeez look at how similar they are?

    I am missing something.

    Predicted is a misnomer. Gravitational wave theorists have been making signal predictions for LIGO under a number of different scenarios for years. One of the most likely signals they expected to see was this type of signal from two black holes orbiting and collapsing into each other. So the overall shape was a prediction, but the exact frequency, chirp rate, and amplitude of the expected signal vary based on the masses of the black holes. So the predicted curves are based on models are then used to fit the data to extract the masses and distance from earth etc.

    If you look at the time scale of the signal it boggles the mind. If I understand it right, each period of the waveform represents an orbital period of the two black holes which is in the millisecond regime. These are objects with masses about 30 times that of our sun. To think that for a brief moment they are revolving around each other faster than the blades of turbo-prop engine at full speed is just crazy.

    • #30
    • February 11, 2016 at 11:59 pm
    • LikeLike
  1. 1
  2. 2
  3. 3
  4. 4