Rumors of Gravitational Wave Inspiral at Advanced LIGO | Sept 2015 Launch

[jedishrfu]

Gravity Wave have reached our shores:

http://www.sciencemag.org/news/2016...nstein-s-ripples-spacetime-spotted-first-time

 

[bcrowell]

Kip Thorne is speaking now. He says the peak power was 50 times greater than the total power output of all the stars in the [observable] universe.

The rumors on the internet were saying there were several events, but at the press conference they're only discussing one. I wonder if the paper will discuss the other events, or if they weren't really statistically significant enough to be publishable.

It's surprising that the signal from the best event is so clearly visible on a time-domain graph, if, as rumored, it was only 5 sigma over all.

 

[maNoFchangE]

Right now the PRL website appears to be overloaded and not responding

Could the majority of the traffic be due to this gravitational wave discovery?

 

[bcrowell]

Since it's CC licensed, it's legal to download and repost it on the internet. If anyone can coax a copy out of their overloaded server and do that, that would be great. I'm still trying and not having much luck.

 

[bcrowell]

Q&A:

One of the things they checked to make sure that it was a real signal was to look at all the data channels for every part of the experiment (servos, etc.), and make sure that none of them was similar to the putative signal.

This version of LIGO is about 3x more sensitive than the previous version. They will tweak the apparatus soon to make it more sensitive by an additional factor of about 3x. The paper will have a statistical analysis of what kind of event rates they expect to see, based on the number of signals they've seen so far in the time it's been operating. The current version is also sensitive to lower frequencies than before.

The concept was first proposed by Russians in 1962.

Unfortunate political compromises, politics of funding. US has not been funding LISA. LISA has released test masses as a test.

 

[Daz]

Phys Rev Lett. Are tweeting snippets of the paper "until their servers are back online." (Obviously overloaded by us lot trying to get a peek!)

https://twitter.com/PhysRevLett

 

[mfb]

It's surprising that the signal from the best event is so clearly visible on a time-domain graph, if, as rumored, it was only 5 sigma over all.

In the CERN seminar this was explained in more detail. They ran for 16 days (of coincidence time between the two detectors) and used this background data to estimate how often stronger signals occur. Their estimate is less than once in ~200.000 years (forgot the second digit). Compare that to 16 days of running time...
The signal is probably more significant, but they would have needed more time for a better background estimate to claim a higher significance.

If the license allows (!) and if someone manages to download the paper, it would be nice to upload it here to help distribution.

 

[websterling]

Looks like the papers are available from LIGO-

https://dcc.ligo.org/LIGO-P150914/public

 

[levadny]

Looks like the papers are available from LIGO-

https://dcc.ligo.org/LIGO-P150914/public

Thanks a lot! I translate in Russian this article.

 

[websterling]

In the CERN seminar this was explained in more detail. They ran for 16 days (of coincidence time between the two detectors) and used this background data to estimate how often stronger signals occur. Their estimate is less than once in ~200.000 years (forgot the second digit). Compare that to 16 days of running time...

The abstract says "a false alarm rate estimated to be less than 1 event per 203 000 years", not the event rate. I saw somewhere that the expected event rate was between 0.5 and several hundred events per year.

 

[sanman]

What major questions of fundamental physics will these Gravity Wave observations help to resolve, and how?

What major questions of astronomy will Gravity Wave observations help to explain?

 

[RJLiberator]

An amazing announcement. I was reading all about these gravitation waves in Brian Greene's books the past few months. Really hits home for me after reading about it.

 

[fresh_42]

The original publication on gravitational waves.

 

[sanman]

Will Gravitational Wave Astronomy help us to verify Dark Matter?

Would it be possible for Gravitational Astronomy to differentiate between Dark Matter and regular matter - perhaps by cross-referencing with other forms of astronomical observation?

 

[RJLiberator]

Great read on the situation: http://www.slate.com/blogs/bad_astr...itational_waves_finally_detected_at_ligo.html

 

[mjs]

Great news-big day for science!
I know its still early, but can someone say which theories are hurt, (and which ones are favored) the most by these new findings?

 

[cockpitvisit]

Was there a corresponding gamma ray burst?

Aren't events that produce gravitational waves (collisions of black holes and neutron stars) also supposed to produce GRBs?

 

[phyzguy]

Great news-big day for science!
I know its still early, but can someone say which theories are hurt, (and which ones are favored) the most by these new findings?

Simple answer is that Einsteinian GR seems to describe the event to near perfection. The paper from LIGO says, "The agreement between the reconstructed waveforms using the two models is found to 94(+2;-3) %". This is the most stringent test yet of GR in the strong gravity regime. I'm not sure if this kills some alternatives to GR, but I think it will disfavor some of them.

 

[PAllen]

Simple answer is that Einsteinian GR seems to describe the event to near perfection. The paper from LIGO says, "The agreement between the reconstructed waveforms using the two models is found to 94(+2;-3) %". This is the most stringent test yet of GR in the strong gravity regime. I'm not sure if this kills some alternatives to GR, but I think it will disfavor some of them.

There is a class of quantum gravity approaches (based on a variant way of handling spin 2 quantization, put forward by a few Russian physicists), that had a specific prediction (yeah!) that collapse would stop signficantly before horizon formation (as I recall, close to the photon sphere of classical GR). Presumably this result (along with increasing reliability of horizon / mass observations) completely kills this approach (the downside of a specific prediction). It probably kills most any approache that suggests the QG effects diverge from classical GR outside the horizon.

 

[atyy]

I really enjoyed Rainer Weiss' explanation, let's hope the press has been paying attention.

Rainer Weiss was one of my experimental physics teachers. I loved his class. I always remember his story about dropping out of MIT. And of course, that he was one of those trying to do this crazy experiment near the quantum limit :)

 

[bcrowell]

It probably kills most any approache that suggests the QG effects diverge from classical GR outside the horizon.

Interesting point. My understanding is that the people saying this kind of thing were working within semiclassical gravity, and they had to renormalize the results of their calculations. It seems to me that once you give yourself the power to arbitrarily renormalize away any effect you feel like getting rid of, you immunize your theory against this kind of straightforward falsification by observation. If they had previously predicted some big effect at or outside the event horizon, now they can probably just say, "Oh, we'll make that go away by subtracting out a certain term from our equations."

 

[PAllen]

Interesting point. My understanding is that the people saying this kind of thing were working within semiclassical gravity, and they had to renormalize the results of their calculations. It seems to me that once you give yourself the power to arbitrarily renormalize away any effect you feel like getting rid of, you immunize your theory against this kind of straightforward falsification by observation. If they had previously predicted some big effect at or outside the event horizon, now they can probably just say, "Oh, we'll make that go away by subtracting out a certain term from our equations."

Yes, but if the claim to distinction of such approach was what it said outside the horizon, it loses that. Unless it has some other point of interest, even its authors might not bother with it anymore.

 

[OrionX76]

Curious if there are other highly sensitive experiments that see an effect from this event. (Such as gama ray detectors or dark matter searches) It would be interesting to take the know arrival time and look for blips in the data...
Or is this effect so weak as to make this search pointless?

 

[marcus]

Dennis Overbye in NYT, 12 Feb:

http://www.nytimes.com/2016/02/12/science/ligo-gravitational-waves-black-holes-einstein.html

 

[.Scott]

The event radiated a total of 3 solar masses in gravitational waves.

What portion of the universe's mass is in the form of gravitational waves?

 

[bcrowell]

What portion of the universe's mass is in the form of gravitational waves?

Almost none, and we don't know why. A maximum-entropy big bang, which is the most overwhelmingly likely possibility, would have had its gravitational degrees of freedom equilibrated with all the other degrees of freedom, so primordial gravitational waves would have been extremely strong. Instead, we got a big bang that was low in entropy, mainly because of the almost complete lack of gravitational waves.

 

[mfb]

That's why I said background ("and used this background data to estimate how often stronger signals occur [by background fluctuations]").
The event rate is poorly constrained with a single signal, of course, but we'll know more in a year or once more events are available (whatever happens faster).

Let's see if LISA gets more funding now. The science case certainly got stronger.

What portion of the universe's mass is in the form of gravitational waves?

Cosmic energy inventory
About 30 part in a billion, with a large uncertainty (~factor 3).

 

[Buzz Bloom]

I noticed some odd inconsistencies between among several reports. (Bolding mine.)

From http://journals.aps.org/prl/pdf/10.1103/PhysRevLett.116.061102

On September 14, 2015 at 09:50:45 UTC the two detectors of the Laser Interferometer Gravitational-Wave
Observatory simultaneously observed a transient gravitational-wave signal​

From https://en.wikipedia.org/wiki/LIGO#Advanced_LIGO

On September 18, 2015, Advanced LIGO began its first formal science observations at about four times the sensitivity of the initial LIGO interferometers.
​

From https://www.ligo.caltech.edu/news/ligo20150918

On, Friday, September 18th 2015, the first official 'observing run' (O1) of LIGO's advanced detectors in Hanford WA and Livingston LA quietly began when the clock struck 8 a.m.​

I looked on the internet for earlier dates than September 18th for when the advanced LIGO began its search for gravitational waves, but that was the earliest date I could find. It seems that the discovery took place four days before the "official" search began.

 

[bcrowell]

It seems that the discovery took place four days before the search began.

This was discussed at the press conference. They were doing an engineering run, which wasn't supposed to be a physics run. However, the apparatus was functioning as it needed to be in order to detect a real signal, and that happened to be when they got this signal.

 

[gentsagree]

Ahah, nice interpretation of the picture.

Yes, they have claimed today they have indeed observed gravitational waves from one event 1.3 billion light years away. The wave form they recorded matches the theoretical prediction, with increasing amplitude and frequency up to a peak, after which it settles down to a more stable configuration. This is indeed what is expected in the event of two black holes orbiting each other at ever decreasing distance, until they merge.

I have a question, which was partially discussed in other places on the web, for anyone to comment on: can gravitational waves be red-shifted?

It seems they can, as they travel against a gravitational potential, from a region of high gravitational field (low potential) to one with lower gravitational field (higher potential). In particular, as electromagnetic radiation, the rate at which time passes as they travel through the potential increases, thus their frequency decreases.

However, as they travel, they also distort time, so I am wondering if in any case these two effects influence each other.

Does anyone have any insight on this?

Thanks!

 

[mfb]

I have a question, which was partially discussed in other places on the web, for anyone to comment on: can gravitational waves be red-shifted?

They are redshifted in the same way light is. The distance gives a redshift of ~9%. The effect of galactic gravitational potentials is negligible.

 

[Buzz Bloom]

the apparatus was functioning as it needed to be in order to detect a real signal, and that happened to be when they got this signal.

Hi bcrowell:

Thanks for the explanation. I did not get a chance to see the press conference.

Of course we don't know how frequently LIGO will find more gravitational waves in the future, but it seems like wonderful luck they had LIGO on the air at that time. The kind of event LIGO detected may well be quite rare, and it LIGO had had its engineering run a day later, it might have been maybe years before any detection occurred. On the other hand, perhaps LIGO will find more BH pair crashes near daily and be overwhelmed with the need to verify so much data.

Regards,
Buzz

 

[Buzz Bloom]

I have a question about the location of the source galaxy.
From http://journals.aps.org/prl/pdf/10.1103/PhysRevLett.116.061102

With only two detectors the source position is primarily determined by the relative arrival Time and localized to an area of approximately 600 deg² (90% credible region).
z = 0.09^+0.03_-0.04.​

If my math is OK, this means that the source can be located with 90% confidence to a region of the sky with an angular diameter of about 14 deg. Combining that with the z value, how many galaxies are candidates for the source?

 

[DiracPool]

What was the eventual fallout from the BICEP2 experiment, btw. Was that totally debunked?

 

[mfb]

I have a question about the location of the source galaxy.
From http://journals.aps.org/prl/pdf/10.1103/PhysRevLett.116.061102

With only two detectors the source position is primarily determined by the relative arrival Time and localized to an area of approximately 600 deg² (90% credible region).
z = 0.09^+0.03_-0.04.​

If my math is OK, this means that the source can be located with 90% confidence to a region of the sky with an angular diameter of about 14 deg. Combining that with the z value, how many galaxies are candidates for the source?

Too many. ~500 million light years uncertainty for the distance, and the distance of ~1.3 billion light years gives ~400 million light years for 14 degrees. So roughly a volume of (400Mly)^3. Tens of millions of galaxies I guess.

What was the eventual fallout from the BICEP2 experiment, btw. Was that totally debunked?

The updated measurement sets an upper limit that excludes the previous value. Gravitational waves could still be there in a sizeable amount, but BICEP2 didn't see them.