SoylentNews is people
SoylentNews
Last month, a team of scientists led by Stacy McGaugh at Case Western Reserve University determined from observations of 153 galaxies that the dynamics of galaxy rotation seems to depend solely on the normal, visible matter in it (SN coverage here). It was a strong argument that rather than hypothesising dark matter to explain the oddities in galactic rotation, it may instead be necessary to modify the laws of gravity.
However, two scientists from McMaster University, Ben Keller and James Wadsley, have just recently examined the results of a detailed simulation of dark matter in galaxy formation previously done known as the McMaster Unbiased Galaxy Simulations 2 (MUGS2). The simulation was a sophisticated one that took into account various other factors such as gas dynamics, star formation, and stellar feedback, but incorporated no new physics beyond that of the standard Lambda-Cold Dark Matter (ΛCDM) cosmological model. They found that the relation that McGaugh et. al. discovered from observations of real galaxies was reproduced just about exactly by the simulation. Their paper is here. Their abstract states:
Recent analysis (McGaugh et al. 2016) of the SPARC galaxy sample found a surprisingly tight relation between the radial acceleration inferred from the rotation curves, and the acceleration due to the baryonic components of the disc. It has been suggested that this relation may be evidence for new physics, beyond ΛCDM. In this letter we show that the 18 galaxies from the MUGS2 match the SPARC acceleration relation. These cosmological simulations of star forming, rotationally supported discs were simulated with a WMAP3 ΛCDM cosmology, and match the SPARC acceleration relation with less scatter than the observational data. These results show that this acceleration law is a consequence of dissipative collapse of baryons, rather than being evidence for exotic dark-sector physics or new dynamical laws.
So now it seems that the earlier troubles with dark matter were actually the result of too naïve a simulation, and by taking into account additional known, relevant physics, the troubles disappear.
Further coverage and commentary by astrophysicist Ethan Siegel here (archive.is).
Related: Study Casts Doubt on Cosmic Acceleration and Dark Energy
(Score: 2) by Bot on Saturday October 29 2016, @10:20AM
Indeed, we do not have observation, hypothesis on interpretation, experiment to prove the hypothesis' predictive power, new theory, repeat.
Here we have Theory, simulation of theory, OK if match with observation.
This way of working is equivalent to
1. the hypothesis of planet to explain some gravitational interference
2. epicycles
in the first case you are OK, in the last case you are not. It is inconclusive.
Account abandoned.
(Score: 3, Interesting) by stormwyrm on Saturday October 29 2016, @02:53PM
There are plenty of branches of science where we cannot perform experiments. Heck, the mechanics of planets in the Solar System is not amenable to controlled experimentation any more than the dynamics of a galaxy is. So since antiquity all astronomers could do was watch and study the planets as they moved across the heavens, and try to figure out how it all worked. They made what amount to simulations of planetary motion by calculating what are called ephemerides. Claudius Ptolemy made such a model of solar system dynamics based on the notions of deferents and epicycles and it took so long for it to be discarded because the Ptolemaic ephemerides matched the crude observations possible with naked eye astronomy closely enough. It was only after the development of the telescope that folks like Galileo and Kepler realised that the Ptolemaic system wasn't good enough, and a better model needed to be made. Kepler made his own laws of planetary motion based on observations, and later Newton came up with a theoretical framework in his laws of motion and universal gravitation that tried to explain why Kepler's laws held.
In the early 19th century astronomers comparing the calculated ephemerides of Uranus with actual observations of the planet realised that its orbit around the sun was perturbed and deviated from the way it should have been moving if it obeyed Newton's laws. They realised that there had to be another planet ("dark matter") beyond Uranus that was causing the perturbations, and soon enough Neptune was discovered. Later that century, better observations of Mercury showed that it too was moving in ways different from what Newtonian mechanics predicted. That led some astronomers (one of them was, in fact the very same astronomer who is credited with discovering Neptune, Urbain le Verrier) to hypothesise yet another planet even closer to the sun, named Vulcan (no relation to Star Trek) that was likewise causing the perturbations. No such planet was ever found, so it took a modification to the theory of gravity in the form of Einstein's theory of General Relativity to explain all the observed features of Mercury's orbit.
I think the parallels to the present-day observations of galactic dynamics and problems in cosmology are rather clear. We have known since the 1930s that the motions of stars in galaxies do not seem to obey the laws of Newtonian dynamics. We hypothesise, just as 19th century astronomers did in hypothesising an unknown planet, that there might be some matter out there surrounding the galaxies which we still cannot see directly that is affecting the motion of the galaxy such that it moves the way it does. Since we can't do an experiment on a galactic scale, the best we can do is make a simulation based on our theories and see how well it matches up with the real stuff out there. Galaxies of course are far more complicated beasts than planets in our solar system, and so observers like Stacy McGaugh discover new properties about them every now and then. When that happens the theorists like Ben Keller and James Wadsley must look at their simulations and see if they are accurate enough to exhibit the features that have been observed. If they do, as has been shown in this case, the theory gets to live a little longer. I don't think it's fair to accuse them of metaphorically "adding more epicycles" in this case: their additions to the simulation were known physics that they already knew would have a measurable effect on the dynamics of galaxies.
The second case though, is a more interesting question. When do we decide that it's time to stop chasing Vulcan and start looking for a new theory of gravitation? Perhaps when attempting to shore up a theory in the face of contrary observation or experiment we should ask if what we are doing is the equivalent of adding more epicycles, i.e. complicating the theory without an observational or experimental basis. By hypothesising dark matter, we have added precisely one such epicycle, and we are trying our damnedest to make that go away. Modifying gravity à la TeVeS though... you should see what their equations look like, and how they try to explain phenomena such as the Bullet Cluster and the peaks in the CMB that dark matter is able to explain so well, and then tell me who is adding more epicycles.
Numquam ponenda est pluralitas sine necessitate.