Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The predictions for the LIGO detection rate are very poor. They're based on a sample of just a handful of binary pulsars observed in our Galaxy, which would produce NS-NS mergers. The BH-BH merger rate is almost totally unconstrained, although it is generally thought to be less than the NS-NS merger rate. So the fact that a BH-BH merger was the first detection, and the fact that it was detected so soon after the sensitivity increases is evidence that the BH-BH merger rate is probably somewhat higher than expected. But we won't know for sure until LIGO detects more events and the rate can be better constrained. Sometimes you do just get lucky.

I should add that there are lots of selection biases and educated guesses in all of this, too. The signal from BH-BH mergers is louder and easier to detect from larger distances. At the same time, NSs are probably more common than BHs, but it's not really clear whether there are more NS-NS binaries than BH-BH binaries because NSs receive kicks from the supernova when they are born but BHs (probably) do not. This may have the effect of blowing apart many nascent NS-NS binaries but leaving the BH-BH binaries intact.



If this events are so rare (that we don't even know how rare they are), how is it possible that they achieved the required certainty (5 sigma)?

I guess you could count one looong wave as a series of one-time events/measurements, but it could as well be a loooong interference.


From the paper: "To account for the search background noise varying across the target signal space, candidate and background events are divided into three search classes based on template length. The right panel of Fig. 4 shows the background for the search class of GW150914. The GW150914 detection- statistic value of ρˆ_c = 23.6 is larger than any background event, so only an upper bound can be placed on its false alarm rate. Across the three search classes this bound is 1 in 203 000 years. This translates to a false alarm probability < 2 × 10^−7, corresponding to 5.1σ. A second, independent matched-filter analysis that uses a different method for estimating the significance of its events [85,86], also detected GW150914 with identical signal parameters and consistent significance" (https://dcc.ligo.org/LIGO-P150914/public). Take a look at Figure 4 as well.

In case you'd like to dig deeper, the 85 and 86 mentioned are:

[85] K. Cannon et al., Astrophys. J. 748, 136 (2012).

[86] S. Privitera, S. R. P. Mohapatra, P. Ajith, K. Cannon, N. Fotopoulos, M. A. Frei, C. Hanna, A. J. Weinstein, and J. T. Whelan, Phys. Rev. D 89, 024003 (2014),


It's not a counting experiment, which makes the calculation of a false positive rate somewhat harder. The key for LIGO is certainly that they saw the signal coincident at two stations, far apart.


This is about detection.

To put it another way, you need a single black swan to prove that black swans exists (to whatever sigma).


Isn't the point though that the gravitational wave observatories are looking specifically for "black swans" rather than just observing swans generally. So when a swan with a lower reflectivity is observed then it now fits the "black swan" profile. Could be just a swan covered in soot; you need more data to show that this swan is always black or that the lower reflectivity wasn't caused by a measuring anomaly, etc.

I may have pushed the analogy too far!


[flagged]


This comment is making the page formatting gross. Those special characters with the strike-throughs make the entire page over-wide, thus requiring horizontal scrolling to read comments.


The browser layout engine should break on the spaces (Chrome does). They are just normal spaces, the combining character should have no effect. You have a bug somewhere.

Also, I cannot edit nor delete it now, so tough luck!

        " "    SPACE	                        Basic Latin
               0x0020
        "̶"     COMBINING LONG STROKE OVERLAY	Combining Diacritical Marks
               0x0336


Gecko doesn't seem to break that line.


The Higgs detection was not a single event but resulted from the statistical analysis of many events.


Thank you for correcting me!


Because of the shape of the event, detection in two places, and more importantly, it matching the signature of the theoretical event extremely closely (especially the ringing at the end)


The detection uncertainty is a separate matter from the predicted rate. Sure, if you had a strong prior that GWs should be detected once in a billion years, then you would want a better detection. But as it is the priors on the detection are pretty weak and this is totally consistent with what is expected.


Both detectors detected the same event at slightly different times


About 7ms apart, which ties in well with the distance between the two detectors.


Others have answered other aspects of this, but as I understand it, it is not the case that we don't know how rare they (BH-BH events) are because they are so rare, we don't know how rare they are, because we don't have a really good model for them. So, we don't know how often we'd expect to detect them, once we had a detector.


I believe the issue is that we don't know of they are rare at all. Perhaps they occur quite frequently. That is the big question.


I recall reading some years ago that gravitational wave would be used to prove multiverse theory. How would that scale compared to bh-bh or ns-ns mergers?

Also, have read today that this discovery backs inflationary theory, how so?

It seems highly unlikely that they could say a specific bh-bh merger was the cause. It seems implied they are triangulating the source, with two detectors?


AFAIK no multiverse theory has yet been put forth that is experimentally testable (even in theory given infinite time, energy etc.) So it's not a proper (falsifiable) scientific theory at present, merely a (in my opinion wild) conjecture.


> no multiverse theory has yet been put forth that is experimentally testable

Just to be clear here; that's because there is no theory for a multiverse. Not yet, anyways. Nobody has put one forth yet. When you hear "multiverse" come out of physicist's mouth, it's because it's a concept indirectly related to other theories. The current popular theory which involves a multiverse is string theory. When string theorists do the math, there is some evidence that a multiverse is possible.

However, that doesn't mean much. Even if string theory was correct and little strings are really the fundamental component of everything in the universe, the multiverse part of string theory could still be wrong. The theory isn't reliant on it, it just doesn't forbid it.

I also wouldn't say that it's entirely untestable. There are a couple things that could be indicative of a multiverse that some physicists have looked for: http://phys.org/news/2010-12-scientists-evidence-universes.h...

The source isn't the greatest, but it shows that we can look at the CMB for indirect evidence. With higher resolution scanning years in the future, such a theory may be testable. I only mention this because the way your comment reads, it sounds like you're saying a multiverse would be inherently untestable.


These are very good points. I do agree that it's possible someone could come up with a testable multiverse-type theory.


How about a theory where spacetime touches upon another spacetime and interacts with it, perhaps through gravity?

Or if spacetime folds back onto itself?


You are suggesting possible future events that would provide evidence for a multiverse. That does not make something a falsifiable theory.

Analogy: it could happen that tomorrow Jesus Christ descends from the heavens and brings the day of reckoning. That would prove Christianity to be true, but the fact that this could happen does not make Christianity a falsifiable theory.

A falsifiable theory is a theory that predicts something that we can (in theory) measure today (possibly requiring infinite resources etc).

Your second example (which is not a multiverse theory at all) is actually a good example of a falsifiable theory. People have calculated [1] that if spacetime was folded back onto itself at even just a single point, it would leave a distinct signature in the cosmic microwave background. We do not observe this signature, so we are pretty sure spacetime does not fold back onto itself.

[1] http://arxiv.org/abs/1108.2842


Ok. What if spacetime folds back onto itself over a very long distance? Wouldn't that be (viewed locally, with our limited instruments) as if another version of spacetime touches upon our version of spacetime?


Physics will always be based on observations. Consciousness is fundamentally hinged TO observation. I'd argue that the way your brain works is more fundamental to reality than the physics causing a Mhz of conduction throughout your synapses. In summary, all bullshit theories are possible in spirit of deceit. For why should good senses be wasted on a cohesive system when the mind is simply a slave to its own devices?


I haven't heard of any credible connection between gravitational waves and multiverse theories.


Is there any idea how far away these black holes were? It would be interesting to know the volume of space it can potentially detect evens.


TFA says "they had heard and recorded the sound of two black holes colliding a billion light-years away" and "1.2 billion years ago".

And from the paper: "The source lies at a luminosity distance of 410+160-180 Mpcc corresponding to a redshift z=0.09+0.03-0.04.". (https://dcc.ligo.org/LIGO-P150914/public) Which corresponds to 1.337+0.522-0.587 billion ly (or between 750.2 million and 1.859 billion ly).


Oops, missed that. Thanks!

Looks like there are roughly three million galaxies within a billion light years. Seems like lots of space for black hole pairs to live in. I suppose over the coming years, these gravity wave observatories will nail down just how common they are.


So that's wayyyyyyy outside our galaxy? Any idea how many galaxies fit into a 1 billion ly sphere around the milky way? I'm guessing a shit ton, which makes the detection of a bh merger seem more realistic to me.

That's some serious range!


It was mentioned during the press conference today that gravitational waves are not affected by interstellar/intergalactic dust the same way light is. In theory, once our detectors are good enough we should be able to use gravitational wave astronomy to peer all the way back to the big bang!


So, we'll finally be able to hear what god said at the start of the universe?


This is a beautiful premise for a mind bending book.


I'm sure He either said "let there be light" or "gee, that's funny...."


I'd prefer either "what does this button do?" or "shit, don't press that!".


Those are what you get when you figure out how to construct a machine that looks to the future and can hear what God says at the end of the universe.


"I clicked print and nothing happened!"


If these waves travel at the speed of light shouldn't the distance and time match up...?


This would be true in a static universe, but, during the 1.2 billion years the waves have been traveling, the universe was experiencing accelerating expansion.

For example, the edge of the observable universe is about 46.5 lightyears away, while the universe is thought to be 13.8 billion years.

https://en.wikipedia.org/wiki/Observable_universe#Misconcept...


is about 46.5 lightyears away

I assume you mean 46.5 billion?


Yeah, thanks! I can't edit it any more, unfortunately.


Not if space time it'self is stretching while the light is in transit.


That's around 1/60th the diameter of the observable universe!


Well, the number of solar-mass black holes in our galaxy is about 10^8. Since black holes form from stars, you can assume the probablity of having binaries is probably related to the probablity of having binary systems in stars, which is high. And the distance to the event is several megaparsecs (much bigger than our galaxy). The fact that they detected two 30 solar mass black holes coalescing 2 days after their sensitivity upgrades says that they almost certainly have had other, less pretty, detections in the few months they've been running their detectors for. Or they should go buy some lottery tickets.


What does that mean in plain English?


Estimates about the frequency of observable events are not yet very good.


Do you know the current best estimate for BH-BH and NS-NS mergers? (Or at least, what they were before this experiment moved the needle).

Interested to know what they were shooting for when they spun this experiment up.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: