This is an update to my October 2017 post on the subject. Here's the link: Sigma 5: Gravity Waves. It's been more than three years. Surely, something has changed. Indeed it has. But before proceeding both backward and forward, let me review the results I reported in my earlier post.
These results were produced by a "gravity wave observatory" called LIGO. For more than a decade LIGO had nothing to report. The reason was simple. It's detector was just not sensitive enough. But it went through several generations of upgrades. That last one (Advanced LIGO) did the trick. The data collection run, tagged "O1" for "Observation run 1", ran from from September of 2015 to January of 2016. It produced three events. Each event was caused by two large black holes spiraling together till they merged.
The observatory was then shut down for minor upgrades. At completion, the O2 run took place. It ran from December 2016 to August 2017. O1 and O2 together eventually resulted in 11 events being detected. When I wrote my post five of them had been reported on. Since then, another round of upgrades has been installed. Upon completion, the O3 run was started. It had to be shut down in the middle so it was informally broken into the O3a run (April 2019 to September 2019) and the O3b run (November 2019 to March 2020).
All together, 56 detection events have been identified. And a third observatory (LIGO is actually two observatories, one located in Washington State, and the other in Louisiana) has been brought online. VIRGO is slightly smaller than the two observatories that combine to make up LIGO, but is sensitive enough to detect many of the same events that LIGO can. With three observatories measuring the same event, its location can be narrowed down to a much smaller slice of the sky. And, in general, more information about the event can be collected.
LIGO is currently shut down so that still more updates can be installed. The O4 run is currently slated to start in June of 2022. And VIRGO is also getting upgraded. And other facilities will be coming online soon. They are scattered all over the globe. There are even plans for "LIGO in space", a LIGO-like instrument that would be bigger than it is practical to go with an earth based observatory. Once those first observations proved that it was possible to detect gravity waves funding has ramped up dramatically.
But that's enough of the present and the future for the moment. Let's go to the past. And let's do it by asking a simple question: what's the speed of light?
It has been possible to make observations spanning distances like ten or twenty miles for millennia. Back then it was obvious that sound traveled at a finite speed. You could observe an action and then note a delay measured in seconds before the sound associated with that action reached you. That made it obvious that, if light was not instantaneous, then it at least traveled much faster than sound.
Reasonably accurate measurements of the speed of sound were successfully made hundreds of years ago. We have had a very accurate estimate of the speed of sound for perhaps two hundred years. And scientists were able to establish that sound worked by oscillating something. Normally, this was air. But it could be water. And the speed of sound through water was higher than that of air. And sound couldn't travel through a vacuum at all. So, scientists have long had a good idea of how sound worked.
And the obvious thing to do was to apply what they knew about sound waves to light waves. If the analogy held then light should have a propagation speed. But what was it? "Fast" just doesn't tell us much. Efforts to determine its speed date back to at least 1629. Experiments with cannons and the like determined that it was too fast to measure using standard methods.
That led to an effort based on astronomy. This effort produced the first measurement that was at least in the ball park. Romer in 1676 made detailed observations of the orbits of the moons of Jupiter. He calculated that in order for the observations to make sense it must take about 22 minutes for light to cross from one side of the Earth's orbit around the Sun to the other. That would have been great if astronomers of the day knew exactly how big that orbit was. They didn't. The best guess put the speed of light at about 140,000 miles per second.
What was important about this is it told scientists just how small the time intervals were that they would need to be able to measure. Say they wanted to measure the propagation time of light over distance of 10 miles. At 140,000 MPS light would take 0.00007 seconds to traverse that distance. A stop watch just wasn't going to cut it.
An early scheme depended on a rapidly rotating wheel with teeth on it. Taking light as particles for the moment to make the explanation simple, arrange for particles of light to be shot past the wheel and along to a distant mirror. There they are reflected back, again past the wheel and on to a detector. If things are arranged such that the light has to travel through the part of the wheel where the teeth are, then it will be blocked when a tooth is in the way but can pass freely when it hits a gap.
Now, spin the wheel at high speed. If the wheel is spinning at just the right speed then a particle of light can pass through one gap between teeth on the wheel, bounce off the distant mirror, and then return just in time to pass through the adjacent gap. This setup allows time to be sliced into very small intervals very accurately. Simple calculations suffice. And a different sized disk or a different rotation speed can used to fine tune the interval to whatever is necessary.
This admittedly inaccurate explanation gives you the idea. The point is that with the proper equipment built along these lines things can be arranged so that the light passes through one slit going and a different slit coming. A wide range of time delays can be accommodated. It is simply a matter of dialing the setup in. Once the right combination of rotation speed and disk/tooth size is found, it is a small step to translate the settings into the speed of light they represent.
And, as a result, Foucault was able to come up with a speed of 298,000 KM/s in 1862. This is very close to the modern value of just under 300,000 KM/s. Others improved his setup and came up with similar values. By 1887 Michelson and Morley were confident that they could measure the speed of light very accurately.
Measuring the speed of light very accurately was of secondary importance to them. Their primary interest was in learning was how fast and in what direction the Earth was moving. To do that they needed to very accurately measure the speed of light.
In normal circumstances sound travels through air. Air is the "medium of transmission", the thing that sound vibrates in order to move. But what was the medium of transmission of light? It had to be something, didn't it?
And there were all those things that weren't the medium of transmission. After all, unlike sound, light can easily travel through a vacuum. And since a vacuum is, by definition nothing, all the usual suspects get immediately eliminated. So, scientists posited the existence of something called the "luminiferous aether".
Assuming something exists just because its existence is convenient is not good enough for scientists. They need actual proof that it does exist. And a good place to start is by trying to measure its properties. And the fundamental property that aether had was its ability to transmit light.
And it was assumed that, everything else being equal, the propagation speed of light in aether was constant. That was true for sound and air.
If you kept air moving at a constant speed and kept its temperature constant, and so on, then the speed of sound through it was constant. Conversely, you could determine some of the attributes of air by measuring the speed of sound through it. For instance, fast moving air would result in a different measured speed of sound than slow moving air.
So, assuming the parallel held, the speed and direction the aether was moving could be inferred from a careful measurement of the speed of light in various directions and at various times. And it was assumed that the aether didn't move. The Earth moved through the aether. So, differences in the speed of light led to different speeds for the aether, which in turn led to a measurement of the speed of the Earth through the aether.
All this was speculation piled upon speculation and scientists knew it. But the measurements were expected to turn up something, even if it wasn't exactly what "aether theory" predicted. And repeated measurements should lead to some ideas about aether theory being discarded and other ideas being confirmed. That was all par for the course. But what everybody agreed on was that careful measurements would turn up differences in the measured speed of light.
After all, it was known that the Earth traveled around the Sun at relatively high speed. And the direction of travel changed with the season. That amount of change alone should have been enough to change the speed of light by a measurable amount. The apparatus had been designed to easily and unambiguously detect changes of this magnitude. If other changes turned up as the measurement process progressed, that would just be a bonus.
The problem is that the Michaelson-Morley experiment turned up the result that no one expected. And this "unexpected result" phenomenon pops up in Science all the time. It is a normal part of science. Scientists expect it to happen regularly. They just don't know when it will happen and when it won't. And what this means is that all those "scientists reject my belief, not because it is wrong, but because it doesn't fit in with what they already believe" arguments are nonsense.
If someone provides hard evidence that current scientific thinking is wrong then scientists change their thinking. That's what scientists were forced to do because Michaelson and Morley got the result they did. No scientist liked the result they got. But other scientists were able to reproduce the result in well conducted experiments. So, scientists had to find a way to live with the result, which they eventually did. Scientists reject "unscientific" beliefs, not because they are unscientific, but because they are not backed by solid evidence.
Scientists have been forced by the results of experiment to reject all kinds of sensible ideas. They have been forced to accept ideas that were far more weird and unnatural and unbelievable than anything an outsiders has thrown at them. Why? Because some well done experiment or observation has forced them to. And the Michaelson-Morley result was one of many instances of this.
The Michaelson-Morley result eventually led Einstein to publish his Special Relativity theory in 1905. Their result had dealt a near-fatal blow to the idea that the luminiferous aether existed. But it wasn't until Special Relativity that scientists bailed completely on it. The theory worked. It also did away with the need for aether to exist at all. The real kick in the pants, however, didn't come until ten years later. Einstein introduced General Relativity in 1915. That's when things got really weird.
In 1905 Einstein had built Special Relativity around the idea that the speed of light is constant. That's pretty weird. In order to make things work the theory demanded that all these other not-light things must stretch and shrink. There were still lots of things that remained unchanging. But still, some things that we had thought were unchanging, changed in these predictable ways in circumstances that Einstein laid out.
Okay. That's lot to buy, but the Michaelson-Morley result demanded some kind of weirdness. And Special relativity weirdness was pretty much the minimum amount of weirdness that would get the job done. The problem is that General Relativity took weirdness to a whole new level. We're now talking bat-shit-crazy weird.
You see, General Relativity requires space itself to stretch and shrink. Space, to put it another way, is the luminiferous aether. And it behaves in many ways like the air that sound travels through. It's what vibrates to transmit gravity.
Newton said "objects in motion tend to continue in that motion". Gravity works by literally warping space. So an object thinks it is continuing to travel in a straight line. But gravity causes space to warp and that caused the "straight line" course of the object to bend, not because the object has changed direction, but because "straight" is no longer straight. Like I said, bat-shit-crazy.
And this "space is wiggly" business means that there are such a thing as "gravity waves", instances of space wiggling because, you know, gravity. And I think you can now understand why I, for one, was not having any of it. I was not convinced that gravity waves even existed even though lots of smart people whom I deeply respected believed that they did. But the LIGO results did me in. They were right and I was wrong.
And I have to admit that I am actually happy that I turned out to be wrong. Because, as I observed three years ago, "[e]very time something previously invisible has become visible, tremendous discoveries have been made". And it is important to understand that the first tremendous discovery has already been made. We now know with absolute certainty that gravity waves exist. That's a tremendous discovery if there ever was one.
Beyond that, we know that General Relativity computations about the characteristics of gravity waves work pretty well. For instance, they get their strength about right. Why not 100% right? Maybe. But we know so little about the events behind the observations at this point that we can't say with certainty. All we know about many events comes from running LIGO data through General Relativity. That results in, for instance an estimated mass. Is the estimate correct? At this point we have no way of knowing for sure.
But even if the calculations are off by some they still tell us things. Remember that first estimate for the sped of light. It was close enough to tell us where the decimal point went. And that was valuable information.
And we now have 56 events to go on. The first event was scary. Was it some kind of screw up? Was it some kind of unusual event or was it pretty typical? With one event it's hard to judge. With 56 events patterns emerge. A lot of the events are two black holes spiraling together to merge into one. We now have some idea of how common that event is. One of the early events was a neutron star merging with a black hole. Scientist got very excited about that one.
There have been some events that fall outside the accepted theories for how these kinds of events are supposed to progress. The details are complex and I don't really understand them. But the scientists are very excited by what they are seeing. It would be nice if everyone else was too. But they are not.
Something the general public doesn't understand is that scientists are actually happier when the data doesn't conform to current theory. It's just more fun and interesting to be on the hunt for a new theory to replace an old broken one. That's as good as it gets. It's what made Einstein famous.
Next best is to come up with a modification to an old theory. Sometimes you don't have to throw the whole thing away. Maybe you change parts of a theory but leave the rest alone. If the result is that the revised version now fits all the experimental data then that is a very good result.
It's progress but not the best outcome if you can change a theory and the new theory is a better but not a perfect fit for the experimental data. That's an improvement, but it also is evidence that more work is needed. Unlike many, scientists expect their theories to agree with all the data, not just most of it.
The scientists that feed off of the LIGO data have gone from not excited to very excited. Before 2016 they had no data to work with. That's not very exciting. Now that they have data, and lots of it, to work with they are very excited.
Unfortunately, things have gone in the other direction in terms of general interest. There was a flurry of press coverage back in 2016. Although the first event LIGO observed happened in 2015 it wasn't announced until then. For reasons I go into in the previous post the first event was suspicious. No one wanted to make an announcement they would later have to take back. So there was a long delay while things were checked and rechecked.
Fortunately, the second event came along pretty quickly. That's when I and many others relaxed. It was real. And a third event followed shortly thereafter. And VIRGO came on line. This was enough to maintain interest until about the summer of 2017. I wrote my post in October of 2017. It turns out that interest by the press and by the public was already waning by then. Press coverage since has been almost nonexistent.
But the data keeps pouring out. The upgrade from the O1 setup to the O2 setup was modest. But it was enough to increase the rate of event detection. The modest upgrade that was sandwiched between the O2 and the O3 runs has also increased the rate of event detection. LIGO will be down for a long time between the O3 and the O4 runs. The currently scheduled starting date for the O4 run envisions a 27 month gap. The gap is so long because the upgrade will be much more extensive.
Each upgrade increases the sensitivity. That means that events similar in size to currently detectable events can be detected further out. Since a "cube" law is involved, a 10% increase in sensitivity translates into a 33% increase in the volume covered. Also, smaller events that happen within the old volume can now be detected. The difference is not as dramatic, but it should result in still more events being detected.
So LIGO started out as what appeared to be a boondoggle. For a long time it ate lots of money while producing no science. But the project did a one-eighty in 2016 when that spectacular discovery of the first event was announced. The discovery of the second event didn't strike the public as nearly as spectacular. But in many ways it was more important. It proved that the first event wasn't a one-off.
Unfortunately, the public saw not much difference between the first event and the second, so they started tuning out. And the public was treating each new event as routine a long time before LIGO got to the 56th one. And routine is not newsworthy. So, the press has been checked out since 2018. It is possible but unlikely that O4 will produce a result that is spectacular enough to put LIGO on the front page again.
That is sad. The quality of the science is increasing by leaps and bounds. A big reason for this is the large pool of events, the very thing that makes the whole enterprise boring to the public. And the O4 run should make things worse at generating buzz by producing data much more quickly than any previous run.
But more data is good for science. Many more events means that comparisons can be made and patterns can be confirmed or disproven. There is lots more data to use to test theories against. Most theories will be found wanting but that's okay. It's how science works.
The practical effect of something as exotic as gravitational waves can not be predicted. No one knew that the time of its development that an obscure and insanely difficult physics theory called Quantum Mechanics would prove to be the foundation upon which all the integrated circuits that power all of our modern electronic devices are built.
Some theoretical work, and at this point LIGO is all about the theoretical, never seems to lead to anything practical. But time after time, something wildly theoretical and of no apparent practical use, ends up allowing us to go from "why are people we don't care about and who live in an obscure corner in China getting sick?" to "we are now making life saving vaccines out of something that the public has never heard of called 'mRNA'." And these mRNA vaccines are so powerful that they can stop a deadly world wide pandemic in its tracks. And only a year separates these two events.
No comments:
Post a Comment