From Megan Gannon at LiveScience:
BERLIN—As expected by many, the 2017 Nobel Prize for physics went to three scientists who helped detect gravitational waves, ripples in space-time predicted by Einstein.
“This year’s prize is about a discovery that shook the world,” said physicist Thors Hans Hansson, announcing the winners from Stockholm.
…
On Sept. 14, 2015, LIGO’s two extremely sensitive instruments in Washington state and Louisiana simultaneously observed a faint gravitational-wave signal. The ripples in space-time came from a pair of two massive black holes that spiraled into each other 1.3 billion years ago. More.
Our physics color commentator Rob Sheldon is, however, left with more questions than answers:
The Nobel was awarded because the waves were finally seen–after 10 years and $1bn of building, looking, upgrading, etc. The waves that were seen, however, are astrophysically impossible, and have been inexplicable for two years, which is enormously long time compared to the speed with which theorists can publish new theories. The first event occurred during engineering testing, and I, among others, expressed skepticism that it looked just like the dry-run simulations. Three more events have been recorded, the last including a third detector in Italy, which pretty much excludes a US conspiracy theory.
So what are they?
If you look at the raw data, and then look at the finished product, you would probably have the same doubts as I. The noise is 1000 times higher than the signal, so 99.95% of the data has to be removed to see the putative waves. The claim is that when two (or three) stations see the same (processed) signal then they know it was real–this is called correlation. A paper three months ago took the noise that was subtracted, and proved that it also showed the same correlation. The implication is that something real is being observed but it doesn’t necessarily look like gravity wave—the “shape” is being dictated by the computer processing of the raw data.
But what would have this correlation?
Earthquakes, ionospheric disturbances, nuclear bomb tests… But there is another one that was detected by accident at the South Pole “Ice Cube” detector. High-energy cosmic rays, zipping through the ice, produce sound-waves and microwaves that then escape the ice and are detected by high-altitude balloons. Most of the “noise” in the LIGO detectors is earthquakes and ground movements. If something of just the right frequency were to make a passage beneath 2 detectors, there would be a correlation of just the right lag. It would even have the right “shape” caused by dispersion. Given the incidence of such events at South Pole–about 1 per day from an area of observation of about 200miles diameter–we can calculate the number that should appear in 2 detectors separated by Livingston-Hanford distance. = (200mi/1800mi)^2 ~ 1/81 * 1/day => 1 per 3 months. Given 4 events in 18 months, this is a pretty close estimate.
Why do I suspect these are not gravity waves?
a) They require mergers of pairs of 30-solar-mass black holes, which have never been seen in isolation (black holes weigh in at either <3 solar masses or > 100,000 solar masses). No one knows how to make a 30 solar mass black hole–best guess is a 100-solar-mass gigantic star, but less than 10 of those per galaxy and we need two, in close proximity?
b) A pair of 30-mass black holes orbiting around each other have virtually no drag, no viscosity. They should orbit for longer than the age of the universe. We don’t know what is slowing them down so as to make them merge.
c) Since stellar-mass black holes are thought to be made in supernova explosions, how can two of them be still gravitationally bound together after two super-supernovae explosions?
d) All four detections have seen the same mass (30) and at the same distance from Earth (500 Megaparsecs). There is no scatter in the data.
e) The resonance frequency of this collision is right around 60Hz–which is also used for lots of other equipment like power lines. It also at the frequency that is “most sensitive” for the two/three detectors. e.g. if there were going to be spurious signals detected, it would appear first in the most sensitive channels. Take a look at the recent Virgo paper results
and especially the power-spectra plots. Remind yourself that 99.95% of the power was removed to create those plots.
Is this worthy of a Nobel prize?
I would say that the Nobel has become nearly completely politicized. Big governments fund these multi-billion dollar efforts, and the Nobel, by its very construction, has to reflect the number of scientists employed and $ spent in the field. Plasma physicists talk about this all the time. Even admitting this bias, two people who made this LIGO effort possible were never rewarded in their lifetime–Joseph Weber and Ronald Drever. In my mind, these were the truly deserving people. So this simply adds another reason alongside the Dark Energy Nobel Prize, why I think the Physics Nobel is no more honorable than the Nobel Peace Prize.
See Dark Energy Wins Nobel Prize in Physics (2011) but its existence remains undetermined.
See also: At Forbes: Gravitational waves detection was all just noise, some researchers say