LIGO Gravitational Wave Detection News

[Wrichik Basu]

TL;DR Summary

On observing signal in their interferometers, LIGO and Virgo Collaborations post a tweet carrying the news. Some (or maybe many, not sure of the number) of us here don't have a Twitter account, so I thought of posting the tweets here to keep everyone updated.

Here is the most recent one:

 

[Greg Bernhardt]

This is different than the big discovery a few years ago?

 

[Ibix]

Yes - this is more probable detections by (now) three instruments. The "big one" was just the first direct detection of gravitational waves.

 

[Wrichik Basu]

On January 6, LIGO and Virgo collaborations announced the discovery of a second binary neutron star merger, labeled GW190425. This is the first confirmed gravitational-wave detection based on data from a single observatory. No electromagnetic counterpart was found. This system is notable for having a total mass that exceeds that of known galactic neutron star binaries.



The paper is on arXiv: https://arxiv.org/abs/2001.01761

The page on the LIGO website: https://www.ligo.org/detections/GW190425.php

 

[Wrichik Basu]

And there has been a signal in the detectors on January 7th.

 

[Tendex]

Yes - this is more probable detections by (now) three instruments. The "big one" was just the first direct detection of gravitational waves.

If you take a look at the event link it says at the top that the detection was by two instruments: H1 and L1 from LIGO.

 

[Tendex]

On January 6, LIGO and Virgo collaborations announced the discovery of a second binary neutron star merger, labeled GW190425. This is the first confirmed gravitational-wave detection based on data from a single observatory. No electromagnetic counterpart was found. This system is notable for having a total mass that exceeds that of known galactic neutron star binaries.

I don't get this. So now it is possible to confirm detections from just one instrument? Wasn't detection based on signal correlation from at least two detection locations?

 

[Wrichik Basu]

I don't get this. So now it is possible to confirm detections from just one instrument? Wasn't detection based on signal correlation from at least two detection locations?

This came to my mind too when I saw the tweet. Maybe they have explained in the paper published in arXiv. Anyways, since the question has come up, I will ask them by replying to their tweet.

 

[Wrichik Basu]

Today LIGO posted this tweet (not in reply to my tweet):


Gravitational-wave physicist Christopher Berry has written about GW190425 in his website:
https://cplberry.com/2020/01/06/gw190425/

Quoting some parts so as to make a geist of the article:

On 25 April 2019 the LIGO–Virgo network observed a signal. This was promptly shared with the world as candidate event S190425z [bonus note]. The initial source classification was as a binary neutron star. This caused a flurry of excitement in the astronomical community [bonus note], as the smashing together of two neutron stars should lead to the emission of light. Unfortunately, the sky localization was HUGE (the initial 90% area wass about a quarter of the sky, and the refined localization provided the next day wasn’t much improvement), and the distance was four times that of GW170817 (meaning that any counterpart would be about 16 times fainter). Covering all this area is almost impossible. No convincing counterpart has been found [bonus note].
...
The localization for GW19045 was so large because LIGO Hanford (LHO) was offline at the time. Only LIGO Livingston (LLO) and Virgo were online. The Livingston detector was about 2.8 times more sensitive than Virgo, so pretty much all the information came from Livingston. I’m looking forward to when we have a larger network of detectors at comparable sensitivity online (we really need three detectors observing for a good localization).
...
Now, the search that identified GW190425 was actually just looking for single-detector events: outliers in the distribution of signal-to-noise ratio and signal-consistency as expected for signals. This was a Good Thing™. While the signal-to-noise ratio in Livingston was 12.9 (pretty darn good), the signal-to-noise ration in Virgo was only 2.5 (pretty meh) [bonus note]. This is below the threshold (signal-to-noise ratio of 4) the search algorithms use to look for coincidences (a threshold is there to cut computational expense: the lower the threshold, the more triggers need to be checked). The Bad Thing™ about GW190425 being found by the single-detector search, and being missed by the usual multiple detector search, is that it is much harder to estimate the false-alarm rate—it’s much harder to rule out the possibility of some unusual noise when you don’t have another detector to cross-reference against. We don’t have a final estimate for the significance yet. The initial estimate was 1 in 69,000 years (which relies on significant extrapolation). What we can be certain of is that this event is a noticeable outlier: across the whole of O1, O2 and the first 50 days of O3, it comes second only to GW170817. In short, we can say that GW190425 is worth betting on, but I’m not sure (yet) how heavily you want to bet.

I’m always cautious of single-detector candidates. If you find a high-mass binary black hole (which would be an extremely short template), or something with extremely high spins (indicating that the templates don’t match unless you push to the bounds of what is physical), I would be suspicious. Here, we do have consistent Virgo data, which is good for backing up what is observed in Livingston. It may be a single-detector detection, but it is a multiple-detector observation. To further reassure ourselves about GW190425, we ran our full set of detection algorithms on the Livingston data to check that they all find similar signals, with reasonable signal-consistency test values. Indeed, they do! The best explanation for the data seems to be a gravitational wave.

(Emphasis added.)

 

[Ibix]

If you take a look at the event link it says at the top that the detection was by two instruments: H1 and L1 from LIGO.

Fair enough - I just meant that we now have three detectors where we only initially had two. I didn't check which were involved.

 

[Tendex]

Fair enough - I just meant that we now have three detectors where we only initially had two. I didn't check which were involved.

Sure, no problem, I just mentioned it for additional info. Actually it's been like this for more than two years, and there's been two detections in which the Virgo instrument participated in 2017.

 

[Tendex]

Today LIGO posted this tweet (not in reply to my tweet):


Gravitational-wave physicist Christopher Berry has written about GW190425 in his website:
https://cplberry.com/2020/01/06/gw190425/

Quoting some parts so as to make a geist of the article:

(Emphasis added.)

Hmm, Berry's words could be interpreted as a certain relaxation in confirmation of detection standards in the future and I wonder why, apparently the projected trend was to be the opposite with the addition of more detectors but if single instrument detections are valid even if that means enlarging the area of localization to a fourth of the sky, it appears the incentive for other countries and teams to join the multidetector search with more detectors is somewhat diminished.

 

[JD_PM]

Some (or maybe many, not sure of the number) of us here don't have a Twitter account, so I thought of posting the tweets here to keep everyone updated.

Wrichik Basu good idea, thank you! :)

 

[PeterDonis]

Berry's words could be interpreted as a certain relaxation in confirmation of detection standards in the future

I don't think so. I think he's just trying to be very clear about what was actually observed and what confidence level there is in various possible explanations. I wish more scientists would take that approach when communicating with the public.

if single instrument detections are valid even if that means enlarging the area of localization to a fourth of the sky, it appears the incentive for other countries and teams to join the multidetector search with more detectors is somewhat diminished

You're treating this as though "detection" were a binary thing--"detected" vs. "not detected". It's not. It's a matter of confidence levels. All sorts of observations occur in these detectors all the time; the question is what caused them. In this case, because of the very large difference in significance of the observations in the two detectors, the confidence level in the hypothesis that the cause of these observations was in fact a gravitational wave is lower. If there were more detectors and more sensitive detectors online, the confidence level would be higher.

 

[Ibix]

In this case, because of the very large difference in significance of the observations in the two detectors, the confidence level in the hypothesis that the cause of these observations was in fact a gravitational wave is lower.

Plus the lack of an identified corresponding optical signal, which we expect to exist (even if we missed it this time!) for a neutron star event.

 

[Tendex]

I don't think so. I think he's just trying to be very clear about what was actually observed and what confidence level there is in various possible explanations. I wish more scientists would take that approach when communicating with the public.
You're treating this as though "detection" were a binary thing--"detected" vs. "not detected". It's not. It's a matter of confidence levels. All sorts of observations occur in these detectors all the time; the question is what caused them. In this case, because of the very large difference in significance of the observations in the two detectors, the confidence level in the hypothesis that the cause of these observations was in fact a gravitational wave is lower. If there were more detectors and more sensitive detectors online, the confidence level would be higher.

It's not a bynary thing but whether one considers an event as a confirmed GW detection or not is a decision with a binary outcome and the important thing is where scientists draw the line of confidence level to present one or the other outcome to the public. My impression from Berry's blog is that he is willing to lower those levels.

 

[Wrichik Basu]

My impresdion from Berry's blig is that he is willing to lower those levels.

That's absolutely not the case. It is clear from his blog that they faced more difficulties while drawing a conclusion on this event compared to other events. In addition, if LIGO was willing to lower the confidence level, they would not have been working so hard to make LIGO India operational.

 

[Tendex]

That's absolutely not the case. It is clear from his blog that they faced more difficulties while drawing a conclusion on this event compared to other events. In addition, if LIGO was willing to lower the confidence level, they would not have been working so hard to make LIGO India operational.

You miss my point, the lowering is in accepting and presenting this particular event with a single detector as a confirmed GW event. I'm not talking about any difficulties in the blog or anything about their efforts about LIGO India that I don't know of. How is this not the case?

 

[Ibix]

It's not a bynary thing but whether one considers an event as a confirmed GW detection or not is a decision with a binary outcome and the important thing is where scientists draw the line of confidence level to present one or the other outcome to the public.

That's certainly one way of looking at it. Another is that there's absolutely nothing wrong with saying "here are a bunch of events we'd bet the farm on being GWs", and "here are a bunch of events we'd bet a month's salary on", down to "the next round's on me if these turn out not to be". Probably phrasing it in more formal Bayesian language.

Scientists should absolutely not be afraid to say "probably", "maybe", and "probably not", as well as "yes" and "no". That blog reads like a "very probably, and maybe more work will up our confidence" to me.

 

[Tendex]

Actually the event is listed as S190425( the prefix s given to events considered not confirmed by LIGO), so maybe I should soften my critical judgement as long as is not yet given the GW qualification. Then again all events since August 2017 have this prefix so maybe there is a change of policy in the last years.
Still I think that giving a 99% chance of being a NS merger to this event is not a correct estimation given the circumstances of having one instrument offline and another one not giving a significant signal from usual noise.

 

[Wrichik Basu]

as long as is not yet given the GW qualification

Maybe you should re-read post #4. Especially the tweet. I believe the above discussion would not have ensued if the event had not yet qualified as a detection. S190425z is GW190425. The paper on arXiv is linked in post #4.

 

[Tendex]

Maybe you should re-read post #4. Especially the tweet. I believe the above discussion would not have ensued if the event had not yet qualified as a detection. S190425z is GW190425. The paper on arXiv is linked in post #4.

All my posts except the last one were in response to that tweet and its qualification of the event there and only make sense under that premise, my last post gave the benefit of the doubt Ibix was asking because I looked at the Wikipedia page on Ligo/Virgo events last edited yesterday and it was still named with prefix S. So you got it all backwards I'm afraid.

 

[Ibix]

Still I think that giving a 99% chance of being a NS merger to this event is not a correct estimation given the circumstances of having one instrument offline and another one not giving a significant signal from usual noise.

Confidence would depend on your prior beliefs about noise sources and modelling of detector responses. I certainly don't know enough about noise and system modelling of the LIGO detectors to comment (note that I haven't read the paper, which might explain, yet).

It occurs to me to wonder what confidence was assigned to the interpretation of GW170817 (the other "neutron star merger" event) based solely on the gravitational wave signature. If it's a lot higher than 99% (which isn't that high in the grand scheme of things) then your concern about the confidence assigned to the interpretation of this event would seem unwarranted without some detailed analysis on your part. If not, we probably need to read some papers.

 

[Tendex]

Confidence would depend on your prior beliefs about noise sources and modelling of detector responses. I certainly don't know enough about noise and system modelling of the LIGO detectors to comment (note that I haven't read the paper, which might explain, yet).

It occurs to me to wonder what confidence was assigned to the interpretation of GW170817 (the other "neutron star merger" event) based solely on the gravitational wave signature. If it's a lot higher than 99% (which isn't that high in the grand scheme of things) then your concern about the confidence assigned to the interpretation of this event would seem unwarranted without some detailed analysis on your part. If not, we probably need to read some papers.

Yes, here it's all mostly about priors. The info about confidence is given by False alarm rates and Signal to noise rates in the papers so it is not easy to extract a probability and all these formulas are based on their own priors and statistical analysis. Both examples of NS mergers you cite have the peculiarity that the Virgo instrument had no detection, in the first case it was conveniently assumed that this was due to the signal falling precisely on Virgo's blind spot which helped reducing the localization area.
What is clear from reading the paper linked by the OP from 4 days ago is that they are now accepting one-instrument detections as valid/confirmed events whereas in the Ligo papers and documentation from years ago(after the first announcement) many pages were written about the essential role of correlating signals by time elapsed at light speed to validate observations.
Might this change have come about to avoid the critique by the Bohr's Institute team in Denmark about noise correlations?

 

[Wrichik Basu]

 

[Wrichik Basu]

Some tweets by physicist Christopher Berry regarding the above event:
https://threadreaderapp.com/thread/1216917328841117696.html
Basically when LIGO announced the news of the signal in their detectors, they had still not confirmed whether this was a real signal, or a glitch. As of now, the event has cleared the initial tests, which is evident from the tweet I posted above in #25.

 

[Wrichik Basu]

Another signal!