Waves of uncertainty
In January of this year, it was reported that the LIGO observatory recorded, possibly the second event of the merger of two neutron stars. This information looks great in the media, but many scientists are beginning to have serious doubts about the reliability of the discoveries of the emerging "gravitic-wave astronomy".
In April 2019, the LIGO detector in Livingston, Louisiana detected a combination of objects located about 520 million light-years from Earth. This observation, made with only one detector, at Hanford, was temporarily disabled, and Virgo did not register the phenomenon, but nevertheless considered it a sufficient signal of the phenomenon.
Signal Analysis GW190425 pointed to the collision of a binary system with a total mass of 3,3 - 3,7 times the mass of the Sun (1). This is clearly larger than the masses commonly observed in binary neutron star systems in the Milky Way, which are between 2,5 and 2,9 solar masses. It has been suggested that the discovery may represent a population of double neutron stars that has not been observed before. Not everyone likes this multiplication of beings beyond necessity.
1. Visualization of the collision of the neutron star GW190425.
Fact is that GW190425 was recorded by a single detector means that scientists were unable to determine the exact location, and there is no observational trace in the electromagnetic range, as in the case of GW170817, the first merger of two neutron stars observed by LIGO (which is also doubtful, but more on that below). It is possible that these were not two neutron stars. Perhaps one of the objects Black hole. Maybe both were. But then they would be smaller black holes than any known black hole, and models for the formation of binary black holes would have to be rebuilt.
There are too many of these models and theories to adapt to. Or perhaps "gravitational wave astronomy" will begin to adapt to the scientific rigor of the old fields of space observation?
Too many false positives
Alexander Unziker (2), a German theoretical physicist and respected popular science writer, wrote on Medium in February that, despite huge expectations, the LIGO and VIRGO (3) gravitational wave detectors showed nothing interesting in a year, except for random false positives. According to the scientist, this raises serious doubts about the method used.
With the 2017 Nobel Prize in Physics awarded to Rainer Weiss, Barry K. Barish, and Kip S. Thorne, the question of whether gravitational waves could be detected seemed to be settled once and for all. The decision of the Nobel Committee concerns extremely strong signal detection GW150914 presented at a press conference in February 2016, and the already mentioned signal GW170817, which was attributed to the merger of two neutron stars, since two other telescopes recorded a converging signal.
Since then, they have entered the official scientific scheme of physics. The discoveries evoked enthusiastic responses, and a new era in astronomy was expected. Gravitational waves were supposed to be a "new window" to the Universe, adding to the arsenal of previously known telescopes and leading to completely new types of observation. Many have compared this discovery to Galileo's 1609 telescope. Even more enthusiastic was the increased sensitivity of gravitational wave detectors. Hopes for dozens of exciting discoveries and detections during the O3 observing cycle that began in April 2019 were high. However, so far, Unziker notes, we have nothing.
To be precise, none of the gravitational wave signals recorded over the past few months have been independently verified. Instead, there was an inexplicably high number of false positives and signals, which were then downgraded. Fifteen events failed the validation test with other telescopes. In addition, 19 signals were removed from the test.
Some of them were initially considered very significant - for example, GW191117j was estimated to be an event with a probability of one in 28 billion years, for GW190822c - one in 5 billion years, and for GW200108v - 1 in 100. years. Considering that the observation period under consideration was not even a whole year, there are a lot of such false positives. There may be something wrong with the signaling method itself, Unziker comments.
The criteria for classifying signals as "errors", in his opinion, are not transparent. It's not just his opinion. Renowned theoretical physicist Sabina Hossenfelder, who has previously pointed out shortcomings in LIGO detector data analysis methods, commented on her blog: “This is giving me a headache, folks. If you don't know why your detector picks up something that doesn't seem to be what you expect, how can you trust it when it sees what you expect?
Error interpretation suggests that there is no systematic procedure for separating actual signals from others, other than to avoid flagrant contradictions with other observations. Unfortunately, as many as 53 cases of "candidate discoveries" have one thing in common - no one except the reporter noticed this.
The media tends to prematurely celebrate LIGO/VIRGO discoveries. When subsequent analyzes and searches for confirmation fail, as it has been for several months, there is no more enthusiasm or correction in the media. In this less effective stage, the media show no interest at all.
Only one detection is certain
According to Unziker, if we have followed the development of the situation since the high-profile opening announcement in 2016, the current doubts should not come as a surprise. The first independent evaluation of the data was carried out by a team at the Niels Bohr Institute in Copenhagen led by Andrew D. Jackson. Their analysis of the data revealed strange correlations in the remaining signals, the origin of which is still unclear, despite the team's claims that all anomalies included. Signals are generated when raw data (after extensive preprocessing and filtering) is compared to so-called templates, i.e. theoretically expected signals from numerical simulations of gravitational waves.
However, when analyzing data, such a procedure is appropriate only when the very existence of the signal is established and its shape is precisely known. Otherwise, pattern analysis is a misleading tool. Jackson made this very effective during the presentation, comparing the procedure to automatic image recognition of car license plates. Yes, there are no problems with accurate reading on a blurry image, but only if all cars passing nearby have license plates of exactly the right size and style. However, if the algorithm were applied to images "in nature", it would recognize the license plate from any bright object with black spots. This is what Unziker thinks can happen to gravitational waves.
3. Network of gravitational wave detectors in the world
There were other doubts about the signal detection methodology. In response to criticism, the Copenhagen group developed a method that uses purely statistical characteristics to detect signals without the use of patterns. When applied, the first incident of September 2015 is still clearly visible in the results, but ... so far only this one. Such a strong gravitational wave can be called "good luck" shortly after the launch of the first detector, but after five years, the lack of further confirmed discoveries begins to cause concern. If there is no statistically significant signal in the next ten years, will there be first sighting of GW150915 still considered real?
Some will say that it was later detection of GW170817, that is, the thermonuclear signal of a binary neutron star, consistent with instrumental observations in the gamma-ray region and optical telescopes. Unfortunately, there are many inconsistencies: the detection of LIGO was not discovered until several hours after other telescopes had noted the signal.
The VIRGO lab, launched only three days earlier, gave no recognizable signal. In addition, there was a network outage at LIGO/VIRGO and ESA on the same day. There were doubts about the compatibility of the signal with a neutron star merger, a very weak optical signal, etc. On the other hand, many scientists who study gravitational waves claim that the direction information obtained by LIGO was much more accurate than the information of the other two telescopes, and they say that the find could not have been accidental.
For Unziker, it is a rather disturbing coincidence that the data for both GW150914 and GW170817, the first events of this kind noted at major press conferences, were obtained under “abnormal” circumstances and could not be reproduced under much better technical conditions at the time measurements of long series.
This leads to news like a supposed supernova explosion (which turned out to be an illusion), unique collision of neutron starsit forces scientists to "rethink years of conventional wisdom" or even a 70-solar black hole, which the LIGO team called too hasty confirmation of their theories.
Unziker warns of a situation in which gravitational wave astronomy will acquire an infamous reputation for providing "invisible" (otherwise) astronomical objects. To prevent this from happening, it offers greater transparency of methods, publication of the templates used, standards of analysis, and setting an expiration date for events that are not independently validated.
