Have we really measured gravitational waves? | Hacker News Reader

Have we really measured gravitational waves?

backreaction.blogspot.com

24 pointsSantosh838 comments

From a year ago: "Danish physicists claim to cast doubt on detection of gravitational waves":

https://arstechnica.com/science/2018/10/danish-physicists-cl...

I am a bit shocked that the work by LIGO is so... unprofessional. How can one give away a Novel prize so easily.

I don't really think that's what happened here (that LIGO's work is unprofessional) and I think the vast majority of scientists believe that LIGO's data (and the analysis) are reliable enough to conclude that they did, indeed, demonstrate the existence of gravity waves to a reasonable level of evidence.

Also remember, they are not only detecting signals, they are detecting same signal at two far away sites that could not be explained by anything originating from Earth.

Maybe I misunderstood Sabine Hossenfelder but she wrote:

> They just throw away data that don’t look like they want it to look.

If I take white noise and an assumption about how my data should look like I can basically prove anything.

If I take two white noise generators I am pretty sure I will still be able to find a signal which more less appears at the same time at both generators.

I hope they did more than just select the data which fits their assumption.

A lot of very interesting information has been published by LIGO and in the popular press (no insult intended, but based on what you're saying I'm assuming you don't have any deep understanding of modern physics experiments).

I wouldn't put too much weight on Hossenfelder's blog posts. She is very angry when things don't go exactly her way (in this case, IIUC, she's basically trying to hold LIGO to an unreasonable standard before accepting proof, and casting some aspersions which only experts in the field would be capable of evaluating).

None taken. I indeed don't have much insight into modern physics experiments. Can you recommend other physicist which give different insights into LIGO or other experiments?

Most of the points made here have been explained for a long time at this point and there doesn't seem to be a good reason to doubt we've discovered gravitational waves.

To go over the points made in the article:

* Sabine complains that the plot in the original LIGO PRL paper was “not found using analysis algorithms”. In reality, the claim of gravitational wave detection was not made using that plot, but with rigorous data analysis which has now been replicated by two independent groups [1]. The only reference I can find for her claim is some New Scientist article and in the LIGO paper it only looks like they applied some filtering to the data to clean it up. That is, they presented real data. While the collaboration should have made it more clear what processing they applied I think this is nitpicking. PRL is a general interest publication and tuning plots for pedagogy is a good thing in this case.

* The blog post also tries to throw doubt at the 2017 multi-messenger signal. First off, the fact that the gravitational wave analysis finished after the gamma ray detection doesn't invalidate the result. They measured gravitational waves with an EM counterpart. Not only that, but followup with a network of optical telescopes was able to locate the galaxy from which the event occurred [2]. These observations and the kilonova aftermath line up closely with the expected optical signal for a binary neutron star merger. The reason why the automatic alert didn't go off is because that alert is triggered by an approximation of the analysis software which does not give the most accurate results. The full analysis takes much longer than the single second scale time delay between the gravitational wave and the gamma ray signals.

* It's also brought up that the collaboration has retracted many of their alerts and that often EM telescopes don't turn up anything. This shouldn't concern anyone because the alerts are based on significant approximations to the analysis to make it fast enough for multi-messenger astronomy. In addition, Sabine doesn't give a fair representation of just how difficult EM detection is. When the alert goes out it isn't telling people to look at just one galaxy, there is a huge error bar on the location with the approximate algorithm. This often means hundreds of galaxies. In addition, up until recently there were very few telescopes actually involved with multi-messenger astronomy. Now that the alert system is open to the public it is expected that more EM discoveries will follow. The reason why the 2017 signal was detected by EM in the first place involved a lot of luck and was due to the fact that the neutron star merger happened to take place in a region of sky with only about 30 galaxies within the error bars. The lack of EM detection is just a testament to how difficult this problem is.

* Again, Sabine tries to throw doubt on the measurement by making scary sounding claims about "glitches" in the data. There is an entire field of experimental physics devoted to detector characterization, not just for LIGO, but for every major experiment out there. "Glitches" as Sabine calls them are nothing new and a normal part of science. She also tries to make the collaboration sound unprofessional by calling out the names they give the glitches which doesn't have anything to do with the science here. The collaboration has good methods of dealing with noise in their detector and these types of problems have already been argued about extensively in literature [3]. The results of those arguments is that gravitational waves are real.

* It' claimed that the collaboration just throws away data they don't like which is simply unsupported.

I have to say that in the past I have appreciated Sabine's posts, especially about the prospects of new high energy colliders. There are real scientific concerns about spending that kind of money on a project that isn't likely to turn up new physics. However, after this article I have lost a lot of respect for her opinions.

[1] https://arstechnica.com/science/2018/12/two-independent-anal...

[2] https://qz.com/1102926/how-ligo-virgo-scientists-tracked-dow...

[3] https://arstechnica.com/science/2018/10/danish-physicists-cl...

Edit because I'm bad at formatting

Reply on news.ycombinator.com