"It's the first time the universe has spoken to us through gravitational waves, up to now we've been deaf to them." -Dave Reitze
No doubt about it: the greatest science advance of 2016 was the end of the century-long wait for the first direct detection of gravitational waves. Not only were we able to detect the inspiral and merger of two black holes from their emission of gravitational waves, we were able to do it more than once. The announcement was a 101-year-after-the-fact confirmation of one of Einstein’s greatest and most unique predictions.
But the real achievement isn’t simply that these detections happened, but what becomes possible. Gravitational wave astronomy is a science in its infancy, but is poised to become rich, varied and to open a whole new window on our understanding of the Universe. This isn’t just the discovery of the year, it’s a new type of science for the 21st century.
A minor correction:
On September 14th, 2015, two black holes over a billion light years away spiraled in and merged together.
You mean the merger was detected on September 14th, 2015. Since it happened over a billion light years away, and gravitational waves travel at the speed of light, the merger itself was hundreds of millions of years ago. As writtten, it implies that the resulting gravitational waves arrived here instantaneously, which I know Ethan did not mean to imply.
But the event happened 14th Sept,2015. In our frame of reference.
Since we're the ones talking, we should be using our frame of reference. Unless you think we need to be writing for any lurkers from a distant galaxy....
No, the detection happened in 2015 in our frame of reference, not the merger. The emission of a signal and the detection of that signal are not the same event. The detection of the CMB from the hot Big Bang happens every day. No one would claim the hot Big Bang is happening all the time in our reference frame. Or, to use an example on the other extreme, if a solar flare was emitted from the sun, and we saw it 8 minutes later when the light from the event reached us, no one would claim that the solar flare 'happened' in our reference frame at the moment we detected it.
Yes, and that's the event. The event happened in our frame of reference 14th Sept, 2015.
In the frame of reference of the black holes, what was our date? Was it the date at which sunlight that hit earth ALSO hit the black holes when they coalesced?
The event happened 14th Sept 2015.
It was from a place that's at a billion light years away, but it is NOW a couple billion light years away. The photon left when the place was a bit over half a billion light years away. But that photon has traveled no time to get here in its frame of reference. AND THAT WILL BE TRUE FOR ALL OBSERVERS.
The only fact we here can trivially calculate is the date at which the event was causally linked to us.
And that's 14th September, 2015.
But Ethan recently discussed YET AGAIN what it means to say some very distant object in an expanding universe is X distance away. There are three ways to measure it, but when it comes to when the event happened, when the light from the event reaches us is an invariant.
"The emission of a signal and the detection of that signal are not the same event. "
Indeed they are different events, but I'm not saying they are. I'm saying that the event occurred in our frame of reference at 14th Sept 2015.
If the object traveled from somewhere 1 billion light years away, then that object was closer than that when the photon left the event there. And now, that location is further than a billion light years away. And even further away when the "knowledge" of our detection could get back to the black holes. So what time did that event happen as far as the source could tell?
But we're here. And the date this happened, in this location, is 14 Sept 2015. And that's a pretty easy call to make for the date.
ESPECIALLY since we don't know precisely how far away this was (like, within 10%). Nor precisely how space expanded over the last billion-ish years to back-calculate another figure.
I don't think you understood my original comment. Of course I agree that if you want to say 'the event' is our detection of the gravitational waves from the black hole merger, then of course it occurred on September 14, 2015. But that is not what Ethan wrote. He wrote that the merger itself happened on that date. To figure out the precise date in our reference frame that the merger happened, we would need to know its precise distance, take into account the expansion of space during the signal's transit time, etc. And like you say, there are different ways of talking about distances for objects so far away. But that is all beside the point - even if one cannot say with precision what date in our reference frame the actual merger happened, one can say with certainty that the merger did not happen in our reference frame on Sept. 14, 2015.
@ketchup #6: Yes, you're correct, but you're engaging in unnecessary, obfuscatory pedantry. We don't know the distance to ANYTHING IN THE FREAKING UNIVERSE with enough precision to assign a specific original date to any observed event.
The universal convention in astronomy, which should be trivially obvious to anyone who has read about any kind of observation (the 1987 supernova, the 1006 supernova which created the Crab nebula, etc.) is to apply the date of observation to the event which generated the observation.
"I don’t think you understood my original comment"
I think you claimed this merely because you don't like not being agreed with.
That I pulled that accusation from my ass because it makes me look better and you worse is COMPLETELY irrelevant. Trust me.
No, I understood it. YOU didn't understand why your claim was irrelevant and also obfuscatory.
And the "concern" that "someone" might think that the light propagated instantly to us is only under a strawman or from someone who isn't going to believe it happened a billion years ago anyway, because they're either a deluded woomancer or a indoctrinated YECer. And neither of those would change their mind if we said "14th September, a billion years ago". ALL that would happen is you'd have to say when we observed it too. THEN explain why it was a billion years ago (and how it was 14th Sept then) and how that was calculated, and which version of "now" you used to work that figure out.
OR we could say it happened 14th September, 2015, and that the object is a billion light years away, IF you feel that it is in any way important (IMO, no, adding that supplies nothing I can use. The date we saw it allows me to look it up. How far away? Not so much).
Further to #7, 1st Sept, 2015, the event was still in our future. It was only on the 15th Sept, 2015 that the event was in our past.
We can say the dinosaurs became extinct 65 million years ago, because that was in our past. The event's information passed us in the past. A long time in the past, but definitely our past. So we can say something about it because it is in our past.
We could NOT say in 2001 that the black holes had merged BECAUSE THAT WAS NOT IN OUR PAST YET. It had not happened.
Traveling long distances in the universe, like looking at distant stars, is traveling through time.
And a common trope for time travel fiction is the "You changed the past, but you haven't changed the past YET, so you still have to go into the past to change the thing you changed in the past otherwise this future doesn't exist" trope.
2001, the event hadn't happened YET.
14th Sept, 2015, the event had just happened.
Since then, it had happened, even if you want to say it was a billion years in the past then. But 2001 it wasn't "14 years less than a billion years in the past" because it wasn't in the past here yet.
Michael Kelsey and Wow,
I am NOT trying to make a precise statement about exactly when the event happened. That you keep trying to explain this to me is why I think perhaps I was not originally understood - If I did not make that clear before, I apologize. I fully understand that it is extremely difficult, or even impossible, to say exactly when the merger actually happened. I am just saying that it is incorrect to say the merger happened on September 14, 2015. Unnecessary pedantry? Perhaps. I will give an example illustrating that it is not always an academic matter to distinguish between the even and the detection.
An example: Say a certain type of explosion occurred at the sun, which is 8 light-minutes from here. This is obviously an astronomical event. Say also that this explosion is known to produce a shock wave that travels toward Earth at 0.5c in our reference frame. When the light from the explosion is detected on Earth, it is 8 minutes after the explosion actually happened. During this time, the shock wave has traveled half-way to Earth. If I insist on saying that the event happened when I detected the light from the explosion, I will think I have 16 minutes before the shock wave hits. But really, I only have 8 minutes. Yes, I know that this is not a real scenario. I am using it only as an example to show that there are measurable differences between between the viewpoints of 'the event happened when we detected it' and 'the event happened before we detected it'.
"I am just saying that it is incorrect to say the merger happened on September 14, 2015."
And I'm telling you that this is wrong. It IS correct to say that, sice that is when the event happened *in this goddamned frame of reference*.
And Mike is telling you that your complaint is FAR more nitpicky than what you claim to be against doing.
We CAN tell when it happened here. And that was 14th Sept 2015.
So you have been told that your complaint is incorrect AND the exact same obfuscation that you claim to be avoiding.
Yet somehow you're trying to say that we're just misunderstanding you?
No, you're not listening.
"An example: Say a certain type of explosion occurred at the sun, which is 8 light-minutes from here. "
Yes, you tried that one.
It doesn't work because there's no difference worth worrying about, and we can calculate how far the sun is anyway.
As an example, if we deducted the nearly 8 minutes from the scenario, WHY WOULD WE? It only makes sense to do so if we want to recalculate for some other observer WHO WE ARE NOT.
Imagine this: we know when we detected it, therefore that's the time it happened.
Fairly self explanatory.
Wow #8 and Michael Kelsey,
I do not dispute the 'universal convention in astronomy'. If Forbes were an astrophysical journal, I would say nothing. But there are two reasons I think it is important to make the distinction in this particular case. First, Ethan is trying to reach a lay audience. I do not think one should assume that this lay audience is universally aware of this 'universal convention'. The second reason is that we are talking about gravitational waves instead of an electromagnetic signal. In all previous astronomical observations in history, it was an electromagnetic signal that was detected. It is reasonable to assume that even a lay audience would understand that these electromagnetic signals travel at the speed of light. However, this is a gravitational wave, and despite what Wow #8 said, there is no 'light' that someone might think propogated instantly. How confident are you that all of Ethan's Forbes readers know that gravitational waves travel at the speed of light?
"I do not dispute the ‘universal convention in astronomy’. "
Then stop saying it's wrong. Because saying it is wrong is disputing the universal convention in astronomy.
" But there are two reasons I think it is important to make the distinction"
Do you remember when you typed out you didn't dispute the universal convention in astronomy? You're disputing it again.
"In all previous astronomical observations in history, it was an electromagnetic signal that was detected"
Irrelevant. Information cannot go faster than the speed of light in a vacuum, and that includes gravity. Which goes at the speed of light in a vacuum.
So what the hell is supposed to be the big difference when they propagate at the same goddamned speed?
"It is reasonable to assume that even a lay audience would understand that these electromagnetic signals travel at the speed of light. "
So stop telling people it happened millions of years before complex life happened on earth, because the same sort of confusion is "How do we know when we weren't even there?"
"However, this is a gravitational wave, and despite what Wow #8 said, there is no ‘light’ that someone might think propogated instantly. "
Go fuck yourself, retard. I never said that.
"How confident are you that all of Ethan’s Forbes readers know that gravitational waves travel at the speed of light?"
If you claim it happened a billion years before we were born, you then have to tell them that this is possible because gravitational waves travel at the speed of fucking LIGHT.
Good job avoiding that fucking problem,moron.
Something that I've wondered about for some time, but can't seem to find much info on is... where are the medium sized black holes?
On one side we have small black holes i.e. 5-50 solar masses, and on the total opposite extreme we have several million to several billion solar mass black holes in the middle of large galaxies. But where is the in-between? several hundred solar masses.. several thousand... hundred thousand solar masses?? A naive way would be too say they all merged.. but regardless of how "deep" we look or how far back in time.. there's like none...
SL, a power law would indicate the smaller black holes are vastly more common than the bigger ones.
But the biggest ones are much easier to see, being stuck in the centre of dense galactic cores.
If 95% of black holes are 5-50 solar masses, we would see mostly black holes of that size. And we'd see a LOT of galactic core blackholes, because there's a lot of galaxies, so 95% of the bigger ones could EASILY be just those.
Meaning that we have 1/4% of mid sized ones.
There is also the mechanism.
A black hole loses MOST of its mass in ejection during the supernova. A 60 solar mass star may form a 6 solar mass black hole. Figure complete ass pull since I used to know the rough loss rate, but this isn't much used in life outside astronomical research on black holes or stellar evolution. Therefore to get to mid size, a black hole has to accrete the extra matter.
And that can take a loooong time
Especially since black holes are messy eaters and spray most of the matter in the swallowed star out to space.
Googling would likely bring you most usefully to an academic book on the subject. It's not likely that there would be much on blogs this deep into the subject, because that would be a citation missed out on, and working scientists don't get paid if they're not generating research papers.
@Sinisa #16: The term you want to search is "intermediate mass black hole" (IMBH). Those are the ones in the 10^2-10^4 solar mass range.
The big "mystery" revealed by LIGO are the 25+ solar mass black holes doing the merging. As Wow (#17) said, the best available (fully 3D) supernova models predict that the remnant should always be below 10 solar masses (even for a 50-100 solar mass progenitor star). So where did the black holes seen by LIGO come from?
Presumably, they formed themselves by mergers of smaller binaries. LIGO isn't quite sensitive enough to see, e.g., a 5+5 merger unless it happens fairly close to us (within the Local Group, I think). But 5+5 -> 9, 9+9 -> 17, 17+17 -> 32 gets you to the right mass range without any new physics.
I'm not sure that accretion (@Wow #17) works for this at all. The Eddington limit means that most black hole accretion leads to strong winds or jets blowing away much of the disk mass, rather than having that mass spiral in. I'm not sure what the best quantitative value is, but it's very small compared to a solar mass per year.
Well, accretion by any method, including merging. But galactic core SMBHs are that big because of various forms of accretion. I mean, stars don't fall in as whole stars. They get ripped apart into a disk, even with million-sun black holes.
There's been a lot of work on HOW they combine, since when I started this, it was still "whacko" to consider there were solar masses bigger than about 60, it was still considered any bigger would collapse before ignition into a star into a black hole.
" 5+5 -> 9, 9+9 -> 17, 17+17 -> 32 "
Hey, can I wisecrack at your maths :-P
Black holes are quite resilient, though. Unlike stars, they don't rip easily, and they don't coalesce fast either, because the energy has to dissipate in gravitational waves, rather than velocity of ejected particles and photons. Which are slightly more effective.
It was still a nice little thing for postgrad work to figure out some of the feasible methods to get really big black holes. Never found out further, though. Every job is with computers, and when you have a job, there's bugger all time for fun. Even (or especially) when that fun comes in scare quotes...
@Wow #20: Heh :-P I did the maths on purpose, using $\rightarrow$ instead of "equals" -- something like 5-10% of the initial net mass of two black holes is lost to gravitational radiation during the merger (it's binding energy, essentially).
See GW150914, where 30+35 -> 62; more specificially (30+3/-4) + (35+5/-3) -> (62+4/-3), with 3.0+-0.5 radiated away. The error bars on the gravitation radiation itself are much smaller than on any of the BH masses, because it can be inferred directly from the signal amplitude.
Two black holes of five solar masses each will give you a final (after ringdown) roughly nine (or 9.5) solar mass BH. I did NOT do a detailed calculation for any of the three cases, just rounded off the fraction.
You're quire right that BH don't merge quickly. For most of the lifetime of a BH binary, they're just like any other orbiting masses (see Earth-Sun, or PSR B19313+16), where the gravitational radiation is extremely small. It's only in the final few minutes or so of the inspiral that most of the energy (and hence our observable signal) is emitted.
Hierarchical assembly is probably the best sensible way to get high mass black holes. However, that model has the problem of the SMBH's we see at very, very high redshift. There wasn't time in just a few hundred million years to get through enough stellar generations for them to assemble. So there are models out there for direct collapse of pure hydrogen-helium clouds into IMBHs, without star formation at all. I don't know enough of the detailed astrophysics to judge how realistic that is.
A hypothesis I liked just for the internal visuals was a large enough cloud collapsing could, if VERY dense (well, relative to gas clouds in space) and VERY uniform, produce large-ish quick black hole stars that blow off the atmospheres of nearby stars and cause some further fusion, seeding some heavy elements *early* in the universe, giving us the slightly higher than calculated prominence of metals (at the time, the improved methods may have solved that oddity) in Population II stars.
And early on, photon pressure from the "bright" universe would help retain thrown atmospheres nearby to seed further nearby collapse.
Necessarily, in my mind it was like a huge (or should that be yuge?) space battle. BOOM! CRASH!!
Ethan did do something a while back on early high-mass SMBHs,but it was on forbes. There may have been some useful newer stuff in that on the subject.
@ Michael & Wow
Thank you for the reference to Intermediate size.. was looking for "medium" size before, and that seems to reference 20-30 Sol.m. BH's, what I considered small (same as 2-3 Sol.m.) when compared to million-billion Sol.masses.
So yes, did find couple of papers, mostly dealing with Xray signals coming from possible candidates, and seems that the current theory is that we do expect to find traces of them slowly moving from i.e. middle of galactic disc towards the center region where they merge in order to explain the several million sol. masses BH there.
I'm guessing with better resolution of microlensing we might find some, instead of just relying on when they are emmiting xrays.