Is the Dark Matter Hypothesis Failing to Explain Galaxy Formation?

In summary, the conversation discusses the failures of the dark matter hypothesis in explaining the observed properties and evolution of galaxies. Computer simulations do not match observations, and dark matter detection experiments have not been successful. The conversation also mentions a potential clue in galaxy properties that may point to a different explanation for the observed anomalies. Despite the complexity of simulating galaxy formation, the evidence suggests that dark matter is not a viable solution and should be treated as a failed hypothesis.
  • #1
Saul
271
4
What comes next if the dark matter hypothesis fails? A hypothesis that is fundamentally incorrect blocks any progress. The entire effort is trying to make a round peg fit in a square hole as opposed to looking for the correct mechanism. What theorist and modeler do is try to change the free parameters to make the model fit what is observed. If the fundamental mechanisms are correct the process has meaning and value. The purpose also of comparing simulations to observations is to kill off incorrect mechanisms.

Computer simulations of galaxy formation with Dark Matter do not match how galaxies are observed to form and do not match the observed properties of galaxies.

The simulations for example have half the observed angular momentum as compared to observations of spiral galaxies and the galaxies that form in simulation are significantly smaller than observed. The problem is dark matter thermalizes the motion of the gas clouds which causes them to clump earlier before forming large galaxies and reducing the angular momentum of the resultant galaxy. In addition the thermalization causes in the simulations a larger galaxy bulge than is observed.

A third problem is how the angular momentum (rotational velocity) changes as one approaches the center of the simulated galaxy as compared to observational data. The computer simulations show dark matter should clump at the galaxy's center which should reduce the total angular momentum (rotational velocity) at the center of the galaxy. This is not observed. The spiral galaxy continues to rotate as one moves to the center of the spiral. This clumping of dark matter in the center of galaxy's also breaks up the bars in spiral galaxy in simulations, which makes it difficult to even form bar, which does not make sense as the observational data indicates spiral bars form and have a long lifetime.

The dark matter detection experiments have not been able to detect dark matter. The point of the dark matter detection is to determine if dark matter does or does not exist.

It is telling that there multiple very fundamental observations that dark matter cannot explain and no one has been able to detect dark matter.

There is currently no viable alternative (MOND has at least as many problems as dark matter) to dark matter, which is curious as there is observational evidence that provides a clue as to what is causing the observational anomalies which dark matter does not in computer simulation explain.

To use an analogy, think of the logical methodology that is used to solve crimes. When there is repeating peculiar evidence at multiple murder scenes, the investigators look for a serial criminal as they expect independent murders to have not have the same peculiar evidence.

This next paper shows eight spiral galaxy properties are interrelate including angular momentum, non random. This is likely a clue to what is really the physical cause of the angular momentum and related anomalies, such the large scale structure anomalies, and the large scale velocity anomalies. (Velocities are higher than the would be expected based on the estimated masses.)

As the authors of the paper note when dark matter was hypothesized there was not large survey data available to test the dark matter hypothesis.
http://arxiv.org/abs/0811.1554http://www.nature.com/nature/journal/v455/n7216/abs/nature07366.html

Galaxies appear (my comment Non random) simpler than expected

Galaxies are complex systems the evolution of which apparently results from the interplay of dynamics, star formation, chemical enrichment and feedback from supernova explosions and supermassive black holes (1). The hierarchical theory of galaxy formation holds that galaxies are assembled from smaller pieces, through numerous mergers of cold dark matter (2, 3, 4). The properties of an individual galaxy should be controlled by six independent parameters including mass, angular momentum, baryon fraction, age and size, as well as by the accidents of its recent haphazard merger history. Here we report that a sample of galaxies that were first detected through their neutral hydrogen radio-frequency emission, and are thus free from optical selection effects (5), shows five independent correlations among six independent observables, despite having a wide range of properties. This implies that the structure of these galaxies must be controlled by a single parameter, although we cannot identify this parameter from our data set. Such a degree of organization appears to be at odds with hierarchical galaxy formation, a central tenet of the cold dark matter model in cosmology (6).
 
Last edited:
Astronomy news on Phys.org
  • #2
I think the main problems is that simulating galaxy formation is very hard. There's lots of complex physics involved, we don't quite know how to handle it.

I mean, we also don't know how to simulate star formation, but that doesn't mean there's a problem with Hydrogen.
 
  • #3
nicksauce said:
I think the main problems is that simulating galaxy formation is very hard. There's lots of complex physics involved, we don't quite know how to handle it.

I mean, we also don't know how to simulate star formation, but that doesn't mean there's a problem with Hydrogen.

Of course we know hydrogen exists and that fusion (hydrogen to helium) creates a great deal of energy. What is the connection with hydrogen and stellar fusion to dark matter and the rotational anomaly of spiral galaxies?

What I have said in my comment is consistent with published papers.

Based on two decades of observation (the more recent and more accurate observational data and computer analysis continues to support the earlier findings) searching for direct evidence of the "dark matter particle" and computer simulations and analysis of what is predicted, to what is observed, is not in agreement. The analysis and the observation supports the assertion that dark matter does not physically exist and if dark matter did exist it could not explain the evolution and morphology of galaxies.

Dark matter appears based on observations and theoretical analysis, in published papers to be an incorrect mechanism.

A failed hypothesis should be treated as a failed hypothesis. Physics has moved on from phlogiston. Physics should and will move on from "Dark Matter". The first step in that process is public criticism of the mechanism in question.
 
  • #4
Please find a citation for me from a published paper on what constraints, from not directly detecting dark matter particles, can be placed. From my understanding, the constraints aren't very significant.Edit: My point, and I know that a lot of people agree, is that computer simulations that show that galaxy formation with dark matter doesn't work isn't evidence against dark matter, it just shows that we don't know how to simulate galaxy formation.
 
Last edited:
  • #5
Dark matter has not passed the first hurdle which is that it physically exists. Dark matter requires the creation of a particle that has the necessary properties which simulations indicate unfortunately is not possible.

The observational tests with increasingly sophisticated experimental apparatus does not support the assertion that dark matter exists. The computer simulation results are irrelevant if the "dark matter" particle does exist. Primarily Zeon100 results are negative for the detection of dark matter.

http://blogs.nature.com/news/thegreatbeyond/2010/05/dark_matter_stays_dark.html
XENON 100 is in a class all its own. It uses around 50 kilograms of a liquid xenon compound that is supposed to emit a flash of light if it gets knocked by a passing dark matter particle. It's the size of the detector that really makes it stand out from the crowd: most dark matter experiments to date measure the size of their detectors in grams rather than kilos. More mass makes an interaction more likely and so even after just 11 days of running XENON 100 has something to say about dark matter.

http://arxiv.org/abs/1005.0380
First Dark Matter Results from the XENON100 Experiment

The XENON100 experiment, in operation at the Laboratori Nazionali del Gran Sasso in Italy, is designed to search for dark matter WIMPs scattering off 62 kg of liquid xenon in an ultra-low background dual-phase time projection chamber. In this letter, we present first dark matter results from the analysis of 11.17 live days of non-blind data, acquired in October and November 2009. In the selected fiducial target of 40 kg, and within the pre-defined signal region, we observe no events and hence exclude spin-independent WIMP-nucleon elastic scattering cross-sections above 3.4 x 10^-44 cm^2 for 55 GeV/c^2 WIMPs at 90% confidence level. Below 20 GeV/c^2, this result challenges the interpretation of the CoGeNT and DAMA signals as being due to spin-independent, elastic, light mass WIMP interactions.

As numerous people have stated the first group to find the dark matter particle will be awarded a Nobel prize. However, if the dark matter particle does not exists how does one prove it does not exist? Will the first team that proves the dark matter particle does not exist be awarded a Nobel Prize?

A competing team are arguing that the ZEON100 finding is not definitive. The competing team of course erroneously announced that they had found evidence of dark matter's existence which was not supported by other teams and which was not replicated.

http://arxiv.org/PS_cache/arxiv/pdf/1005/1005.0838v2.pdf
 
  • #6
"and hence exclude spin-independent WIMP-nucleon elastic scattering cross-sections above 3.4 x 10^-44 cm^2 for 55 GeV/c^2 WIMPs at 90% confidence level"

Isn't nearly convincing enough to conclude that dark matter is a failed hypothesis, as you say.
 
  • #7
As I said, the simulations with dark matter produce galaxies with half the angular momentum of real physical galaxies and galaxies that are an order of magnitude smaller than observed galaxies.

The authors of this paper acknowledge the problem and then propose unrealistic heating and unrealistic lack of cooling of the massive gas clouds to try to rectify the problem. What they are proposing does not agree with observations and analysis of how intergalactic gas cools. As they note in the paper that theoretical toy model assumption creates other problems which does not agree with observations.

http://arxiv.org/PS_cache/astro-ph/pdf/0308/0308117v1.pdf


The Angular Momentum of Gas in Proto-Galaxies: II – The Impact of Preheating by
Frank C. van den Bosch, Tom Abel, and Lars Hernquist


Foremost, hydrodynamical simulations of disk formation that include cooling indicate that, contrary to the standard assumption, the specific angular momentum distribution of the gas is not conserved. Instead, the gas looses a large fraction of its angular momentum to the dark matter (Navarro & Benz 1991; Navarro & White 1994), yielding disks that are an order of magnitude too small. This problem has become known as the “angular momentum catastrophe”, and is typically associated with the well-known “over-cooling problem” in CDM cosmologies (White & Rees 1978; White & Frenk 1991). At early times gas cooling is very efficient, leading to the formation of dense gas clumps which loose their orbital angular momentum to the surrounding dark matter haloes through dynamical friction, before eventually merging to form the central disk. Therefore, some mechanism is required to prevent or delay the cooling of the gas, so that it can preserve a larger fraction of its angular momentum. Indeed, simulations in which gas cooling is artificially suppressed until z = 1 yield larger, more realistic disks (Weil, Eke & Efstathiou 1998; Eke, Efstathiou & Wright 2000).

In an ongoing attempt to improve our understanding of the origin and evolution of the angular momentum of baryons, we have performed a number of simulations of structure formation in a LCDM cosmology without cooling. While unrealistic, the absence of cooling allows us to better focus attention on the impact of gravity and (shock) heating on the angular momentum of the baryons. Therefore, such simulations are a logical preliminary step to investigate the build-up of angular momentum in proto-galaxies.

5 DISCUSSION & CONCLUSIONS
A full understanding of the structure and formation of disk galaxies within a hierarchical, cold dark matter cosmogony faces a number of intriguing challenges. First, in the absence of any heating, baryons cool extremely efficiently, producing too many satellite galaxies and resulting in what has become known as the angular momentum catastrophe. In addition, the angular momentum distributions of both the gas and the dark matter in numerical simulations reveal an excess of low angular momentum material compared to real disk galaxies.

However, preheating also has some effects that are less favourable for disk formation. First of all, the total specific angular momentum of the gas within the virial radius, and which is thus eligible to cool and form a disk galaxy, is reduced with respect to that of the dark matter. Thus, although the angular momentum loss may be reduced, there is less angular momentum to start with. Second, the detailed angular momentum distributions reveal a clear increase of the baryonic mass fractions with negative specific angular momentum, making the formation of a disk dominated galaxy less plausible. We therefore conclude that understanding disk formation remains an intriguing puzzle, even in a preheated IGM (intergalactic material).
 
  • #8
If "Dark Matter" does not exist and even if dark matter did exist it could not explain the observations, at what point does astrophysics abandon the dark matter mechanism?

The galaxy rotational anomaly, the galaxy velocity anomaly, the spiral galaxy morophological anomalies (there are obvious unexplained patterns and processes going on) has a physical explanation. What is interesting is the physical explanation for those anomalies appears based on published papers to not be dark matter and is not MOND.

http://arxiv.org/abs/astro-ph/0012334v1


“Dark Halo and Disk Galaxy Scaling Laws” by J. Navarro.


Normal matter interacts gravitational with “dark matter”, so dark matter can loss or gain energy from the galaxy. Unfortunately for the “dark matter theory”, hydro-dynamic simulations, fundamentally disagree with real galaxies. The simulations create a model disc that is an order of magnitude smaller than what is observed. This discovery, which is called the “angular momentum catastrophe”, was made 8 years ago. There is no solution to the angular momentum catastrophe, which is not surprising; however, as more detail data and observations concerning spiral galaxies shows structures that could not possibly have been created by the interaction of “dark matter” and normal matter.

It should be noted that the ‘angular momentum catastrophe” problem and the “missing satellites problem” is leading some researchers to state that dark matter does not exist which is interesting as LCDM theory will need to change. The “angular momentum catastrophe” and the missing satellites problem” are not the only fundamental disagreements with the “dark matter” theory and reality.
 
  • #9
The direct detection attempts to find evidence that dark matter exist have been negative.

Even if dark matter exists, it cannot explain galaxy formation. It seems logically that dark matter does not exist based on the negative results to observe it.

If dark matter existed galaxies would be an order of magnitude smaller and would have half the angular momentum than this universe and its galaxies. As this is not observed, it appears "dark matter" does not exist.


http://arxiv.org/abs/astro-ph/0505226v1

“The mass of dwarf spheroidal galaxies and the missing satellite problem” by Read, Wilkinson, Evans, Gilmore, & Kleyna


3. Conclusions
In conclusion, tidal stripping cannot be very strong for many, if not all, of the local group dSphs. Strong tidal stripping, which would produce distorted isodensity contours, also leads to velocity gradients and flat or rising projected velocity dispersions - neither of which are observed in the local group dSphs for which we have good kinematic data (but see also Munoz et al. 2005). This suggests that dSph galaxies must be sufficiently massive such that tidal stripping is of little importance for the stars. Either they are on orbits with large pericentres, in which case they can have masses as low as _ 10^8M⊙ (Kleyna et al. 2001); or they are on more extreme orbits in which case they must be _ 10^9 − 10^10M⊙ depending on the extremity of the orbit. Our current cosmological paradigm would favour the latter hypothesis, but this leaves us with a puzzle: if the dSph are really as massive as _ 10^10M⊙ and have dark matter densities which are cosmologically consistent then they would have central velocity dispersions which are too large to be consistent with Draco or UMi - even after significant tidal stripping and shocking.
 
  • #10
Unfortunately, galaxy formation is extremely difficult to simulate and the failure to simulate a galaxy from first principles cannot be automatically blamed on the dark matter hypothesis. This isn't because the dark matter itself is difficult to simulate (this part is quite easy), but rather because of uncertainties in the baryonic physics. In a dark-matter dominated universe, we expect galaxies to form in a bottom-up fashion, meaning that little things merge together to form bigger things. To simulate the formation of, for example, the Milky Way, we need to simulate the gas physics, star formation, and AGN and stellar feedback of these little things as they merge together to form bigger things. This means that we need to start our simulations on very small scales and run the simulation until the little things merge together to form the present-day Milky Way (which will be hundreds or thousands of times larger than its original components). Such simulations take tremendous computational resources and even then require us to make a great many approximations.

Direct detection experiments are a more promising avenue for examining the dark matter problem, but unfortunately, we've explored only a very small subset of the parameter space of possible dark matter particles. Even if the dark matter hypothesis is wrong, it will be difficult to rule it out as a possibility anytime in the near future. However, physicists continue to examine possible alternative theories of gravity. If one of these theories were to make a successful prediction that ran contrary to most conventional dark matter theories, I think the community would take notice. Unless that happens, astronomers will continue to work within the dark matter paradigm.

Admittedly, dark matter is difficult to falsify, but that doesn't make it wrong. Rest assured, however, that we are working hard on this problem and you will know as soon as we have hard evidence one way or the other.
 
  • #11
Saul said:
“Dark Halo and Disk Galaxy Scaling Laws” by J. Navarro.

Normal matter interacts gravitational with “dark matter”, so dark matter can loss or gain energy from the galaxy. Unfortunately for the “dark matter theory”, hydro-dynamic simulations, fundamentally disagree with real galaxies. The simulations create a model disc that is an order of magnitude smaller than what is observed. This discovery, which is called the “angular momentum catastrophe”, was made 8 years ago. There is no solution to the angular momentum catastrophe, which is not surprising; however, as more detail data and observations concerning spiral galaxies shows structures that could not possibly have been created by the interaction of “dark matter” and normal matter.

I don't see that anywhere in Julio's paper. Where are you quoting from?
 
  • #12
It took decades to detect neutrinos after Pauli hypothesized their existence, so, it is unsurprising it is taking time to detect dark matter - a much more difficult endeavor. The evidence in favor of dark matter is overwhelming [e.g., bullet cluster] in the minds of most mainstream scientists. Disney has a history of flamboyant opposition to mainstream ideas, so I would suggest a grain of salt when reading his papers.
 
  • #13
SpaceTiger said:
Admittedly, dark matter is difficult to falsify, but that doesn't make it wrong.

Same happens with the Tooth Fairy. Rest assured though we are working hard on it.
 
  • #14
AWA said:
Same happens with the Tooth Fairy. Rest assured though we are working hard on it.

Might I suggest putting a camera in your bedroom. :smile:
 
  • #15
SpaceTiger said:
Unfortunately, galaxy formation is extremely difficult to simulate and the failure to simulate a galaxy from first principles cannot be automatically blamed on the dark matter hypothesis. This isn't because the dark matter itself is difficult to simulate (this part is quite easy), but rather because of uncertainties in the baryonic physics. In a dark-matter dominated universe, we expect galaxies to form in a bottom-up fashion, meaning that little things merge together to form bigger things. To simulate the formation of, for example, the Milky Way, we need to simulate the gas physics, star formation, and AGN and stellar feedback of these little things as they merge together to form bigger things. This means that we need to start our simulations on very small scales and run the simulation until the little things merge together to form the present-day Milky Way (which will be hundreds or thousands of times larger than its original components). Such simulations take tremendous computational resources and even then require us to make a great many approximations.

Spiral galaxies show indicates of patterns which indicates they were not formed from many mergers.

A second problem with the many mergers hypothesis is that simulations indicate multiple mergers of spiral with spiral produces an elliptical galaxy. Observationally that is not observed. The number of spiral galaxies stays roughly constant with redshift at 70%.

It should be noted that the paradox between simulation properties and observed properties are fundamental properties of the spiral. Angular momentum and change in angular momentum as one moves towards the core of the galaxy for example.

Saul's comment:Normal matter interacts gravitational with “dark matter”, so dark matter can loss or gain energy from the galaxy. Unfortunately for the “dark matter theory”, hydro-dynamic simulations, fundamentally disagree with real galaxies. The simulations create a model disc that is an order of magnitude smaller than what is observed. This discovery, which is called the “angular momentum catastrophe”, was made 8 years ago. There is no solution to the angular momentum catastrophe, which is not surprising; however, as more detail data and observations concerning spiral galaxies shows structures that could not possibly have been created by the interaction of “dark matter” and normal matter.

Space Tiger: I don't see that anywhere in Julio's paper. Where are you quoting from?
Saul: My error, I copied and quoted one of my own comments. That specific comment is the issue.

http://arxiv.org/abs/0811.1554

http://www.nature.com/nature/journal...ture07366.html

Galaxies appear (my comment Non random) simpler than expected

Galaxies are complex systems the evolution of which apparently results from the interplay of dynamics, star formation, chemical enrichment and feedback from supernova explosions and supermassive black holes (1). The hierarchical theory of galaxy formation holds that galaxies are assembled from smaller pieces, through numerous mergers of cold dark matter (2, 3, 4). The properties of an individual galaxy should be controlled by six independent parameters including mass, angular momentum, baryon fraction, age and size, as well as by the accidents of its recent haphazard merger history. Here we report that a sample of galaxies that were first detected through their neutral hydrogen radio-frequency emission, and are thus free from optical selection effects (5), shows five independent correlations among six independent observables, despite having a wide range of properties. This implies that the structure of these galaxies must be controlled by a single parameter, although we cannot identify this parameter from our data set. Such a degree of organization appears to be at odds with hierarchical galaxy formation, a central tenet of the cold dark matter model in cosmology (6).
 
Last edited by a moderator:
  • #16
SpaceTiger said:
Might I suggest putting a camera in your bedroom. :smile:

You got me there :smile:
 
  • #17
As noted in my above comments, the dark matter hypothesis does not explain spiral galaxy morphology or evolution. There are multiple specific significant gaps between observation and simulation. For example the simulation spiral galaxies have half the angular momentum and are smaller than physical spiral galaxies.

The claim that there is a gap between simulation and observation due to the complexity of the of performing a many body simulation, misses the issue or more appropriate the problem situation. The observational data is pointing to a difficult physical cause.

A different observational problem that I noted in my above comment, is mergers of spiral galaxies with spiral galaxies should produce an elliptical galaxy. The observation that the percentage of spiral galaxies does not vary with redshift and remains at 70% is interesting as it is a clue to what causing the rotational anomaly in spiral galaxies and the systematic morphology changes in spiral galaxies. A second related question is what is causing the difference between spiral and elliptical galaxies? (i.e. What is causing the two very different galaxy types to form.)

An elliptical galaxy's stars do not all rotate about the galaxy's gravitational center in the same direction. That observational fact makes sense based on conservation of angular momentum and the random direction of merging galaxies. What one would expect, if elliptical galaxies formed from multiple mergers of gas clouds or other galaxies is the elliptical would have a small net rotation (its stars rotate in both directions and hence the angular momentum cancels) and should have multiple rotational axises which is what is observed.

The reason that there has been no theoretical progress to explain the spiral galaxy observations is the initial proposed hypothesis "dark matter" was proposed at a time when there has not very accurate survey data showing how galaxies evolve with redshift.

http://arxiv.org/abs/astro-ph/0702585v1

“The Milky Way: An Exceptionally Quiet Galaxy; Implications for the formation of spiral galaxies”, by F. Hammer, M. Puech, L. Chemin, H. Flores, M. Lehnert

Disk galaxies constitute the majority of the galaxy population observed in the local universe. They represent 70% of intermediate mass galaxies (stellar masses ranging from 3× 10^10 to 3 × 10^11 M⊙), which themselves include at least two-third of the present-day stellar mass (e.g., Hammer et al. 2005). Early studies of the Milky Way have led to a general description of the formation of a disk galaxy embedded in a halo (Eggen, Lynden-Bell, & Sandage 1962). Fall & Efstathiou (1980) set out a model of galaxy formation assuming that disks form from gas cooling and condensing in dark halos. Protogalactic disks are assumed to be made of gas containing substantial amount of angular momentum, which condenses into stars to form thin disks (Larson 1976). These disks then evolve only through secular processes.

However, there are several outstanding difficulties with this standard scenario. One such difficulty is the so-called angular momentum problem. That is, simulated galaxies cannot reproduce the large angular momentum observed in nearby spiral galaxies (e.g., Steinmetz & Navarro 1999). Another is the assumed absence of collisions during and after the gas condensation process. Indeed, the hierarchical nature of the _CDM cosmology predicts that galaxies have assembled a significant fraction of their masses through collisions with other galaxies. It is likely that such collisions would easily destroy galactic disks (e.g., Toth & Ostriker 1992).

Using arguments based on either dynamical friction (Binney & Tremaine 1987) or simple orbital time-scale (e.g.,Bell et al. 2006), this time scale has been estimated to be about 0.35Gyrs. Combining the pair fraction and characteristic time scale estimates suggests that for a present-day galaxy with a stellar mass larger than 3 × 10^10 M⊙, the chance it has experienced a major merger since z=1 is 50±17%, 75±25% and 70% according to Lotz et al. (2006), Hammer et al. (2005), and Bell et al. (2006), respectively1. Although less certain, integrating the merger rate to higher redshift implies that a typical bright galaxy may have experienced up to four to five major merging events since z=3 (Conselice et al. 2003).

The widely accepted assumption that a major merger would unavoidably lead to an elliptical is perhaps no longer tenable: accounting for the large number of major mergers that have apparently occurred since z=3 would imply that all present day galaxies should be ellipticals. This is obviously not the case. So it is likely that disks either can survive or are “rebuilt” after a major merger, through whatever mechanism as yet perhaps unknown in detail (see, for example, Robertson et al. 2006).
 
  • #18
Saul do you have any working hypothesis that might substitute the "dark matter" hypothesis for the shape of the spiral galaxies?
 
  • #19
Chronos said:
It took decades to detect neutrinos after Pauli hypothesized their existence, so, it is unsurprising it is taking time to detect dark matter - a much more difficult endeavor. The evidence in favor of dark matter is overwhelming [e.g., bullet cluster] in the minds of most mainstream scientists. Disney has a history of flamboyant opposition to mainstream ideas, so I would suggest a grain of salt when reading his papers.

From a theoretical standpoint, the neutrino is required from energy/mass balance. (i.e. The point is there is a clear unequivocal theoretical requirement which indicate that the "neutrino" must exist.)

It does not follow from the fact there is a clear theoretical requirement that neutrinos exist and that the neutrino was found that there is a theoretical requirement for "dark matter" and that the "dark matter" exists.

Science is the proof and disproof of hypotheses. For example science move beyond phlogiston.

http://en.wikipedia.org/wiki/Phlogiston_theory

There are degrees of certainty concerning different hypotheses.

The fact that no one has been able to detect dark matter using the most sophisticated detection apparatus and that there are multiple fundamental observations which the dark matter hypothesis cannot explain, justify the criticism of the theory.

Dark matter has specific predicted theoretical affects on rotational speeds, galaxy formation, and galaxy morphology.

As noted above those specific dark matter theoretical predictions have been shown to not be correct. In the normal process of science we would agree the theory has been falsified and the field is in a crisis. (i.e. There are multiple fundamental observations that have no explanation and the observations in question appear to be connected.)

In other fields of science the participants would not propose changing the laws of physics or to create a new particle to explain observations.
 
  • #20
As I see it, the evidence for Dark Matter consists entirely of the fact that matter on scales larger than the solar system gravitates more strongly than predicted by GR, which means that if we accept GR any extra gravitation must therefore be caused by invisible matter. The evidence is therefore equivalent to the evidence that GR is correct in this regime.

That evidence in turn consists mainly of the fact that GR is a neat and self-consistent theory and that it passes solar system experimental tests accurately, where many proposed alternatives, often more complex, have failed to do so.

However, I personally don't think GR is completely right, both on theoretical and experimental grounds. I think it's probably a very accurate local weak-field approximation to a different theory which gives very different results on a larger scale. I'm looking forward to seeing more experimental results to help give clues as to the nature of that theory. Unfortunately, these forums are not the place to speculate about such things.
 
  • #21
Jonathan Scott said:
As I see it, the evidence for Dark Matter consists entirely of the fact that matter on scales larger than the solar system gravitates more strongly than predicted by GR, which means that if we accept GR any extra gravitation must therefore be caused by invisible matter. The evidence is therefore equivalent to the evidence that GR is correct in this regime.

However, I personally don't think GR is completely right, both on theoretical and experimental grounds. I think it's probably a very accurate local weak-field approximation to a different theory which gives very different results on a larger scale. I'm looking forward to seeing more experimental results to help give clues as to the nature of that theory. Unfortunately, these forums are not the place to speculate about such things.

Jonathan,

Try to explain the observations. What you are suggesting is substituting one incorrect hypothesis with another incorrect hypothesis. Read Disney et al's paper. Think about the Tully-Fisher relationship which is a key dependent parameter but only one of the dependent parameters. There is a pattern in the spiral galaxy observations. Why?

Disney's paper lays out part of the problem situation. The observational anomalies are not just that the angular momentum of spiral galaxies is higher than can be explained by the estimated mass of the galaxy and not just that the velocity of galactic clusters is higher than expected.

Everyone seems to have some mental block to consider other possible explanations to what is observed and to use alternative methodologies to problem solve. What are the field's fundamental assumptions? Could any of the fundamental assumptions be incorrect based on the observations? What are the other observational anomalies? The methodology people use to approach this problem is not effective. Trying to fit a round peg in a square hole for years. Part of the scientific process is identifying that there is a round peg and a square hole.

Trying to make a failed theory work does advance the process. It actually appears to block any progress. Dark matter appears obviously to be on the ropes. Could it make a come back? It appears very, very, unlikely.

The theory must be on the correct page. When is it clear the theory is not on the correct page, people appear to have no idea how to rationally move beyond the blockage. The observations are clues. Stop guessing. Discuss the observations from a higher level.

http://arxiv.org/ftp/arxiv/papers/0811/0811.1554.pdf

Galaxies appear simpler than expected
M. J. Disney1, J. D. Romano1,2, D. A. Garcia-Appadoo3,1, A. A. West4,5, J. J. Dalcanton5 & L. Cortese1

The rogue colour is scattered more or less randomly, and is correlated with none of the other observables. (It would be natural, although not compelling, to identify the systematic colour with the old population of stars, and the rogue component with recent and transitory bursts of star formation—very luminous young stars being far bluer and shorter lived than older ones.) The remaining six 5 variables, including the systematic colour, are all correlated with one another, and with a single principal component. In other words they form a one-parameter set lying on a single fundamental line. Such simplicity was unpredicted and is noteworthy when galaxies might well have been controlled by any six of the seven potentially independent physical parameters listed earlier. It is even more significant considering the enormous variety amongst the galaxies in the sample11.

If, as we have argued, galaxies come from at most a six-parameter set, then for gaseous galaxies to appear as a one-parameter set, as observed here, the theory of galaxy formation and evolution must supply five independent constraint equations to constrain the observations. This is such a stringent set of requirements that it is hard to imagine any theory, apart from the correct one, fulfilling them all. For instance, consider heirarchical galaxy formation in the dark matter model, which has been widely discussed in the literature3,4. Even after extensive simplification, it still contains four parameters per galaxy: mass, spin, halo-concentration index and epoch of formation. Consider spin alone, which is thought to be the result of early tidal torquing. Simulations produce spins, independent of mass, with a log-normal distribution. Higher-spin discs naturally cannot contract as far; thus, to a much greater extent than for low-spin discs, their dynamics is controlled by their dark halos, so it is unexpected to see the nearly constant dynamical-mass/luminosity ratio that we and others14 actually observe. Heirarchical galaxy formation simply does not fit the constraints set by the correlation structure in the Equatorial Survey.

More generally, a process of hierarchical merging, in which the present properties of any galaxy are determined by the necessarily haphazard details of its last major mergers, hardly seems consistent with the very high degree of organisation revealed in this analysis. Hierarchical galaxy formation does not explain the commonplace gaseous galaxies we observe. So much organization, and a single controlling parameter—which cannot be identified for now—argue for some simpler model of formation. It would be illuminating to identify the single controlling galaxy parameter, but this cannot be attempted from the present data.

It is natural to ask why this fundamental line was not discovered before. To some extent it was because even the pioneers24,25 and others26,27,28 working with small numbers of optically selected spirals could reduce six observables to two; one relating to size, one to morphology. The strong optical selection effects, which hamper optical astronomers in detecting and measuring galaxies whose surface brightnesses are barely brighter than the sky5,29, disguised galaxies’ simplicity.
 
Last edited:
  • #22
Disney et al's paper "Galaxies appear simpler than expected" http://arxiv.org/ftp/arxiv/papers/0811/0811.1554.pdf referenced this paper which is a review paper on hierarchical galaxy formation. Disney's et al's analysis shows patterned non random spiral properties which contradicts the hierarchical galaxy paradigm and indicates there is an unknown parameter which is controlling these connected parameters.

Some of the observations that the Disney paper shows are correlated are included as unexplained anomalies in the review paper

The Tully-Fisher relationship between disk galaxy luminosity and the disk galaxy's rotational speed for example. The point is a disk galaxy's rotational speed is in the hierarchical galaxy paradigm due to random mergers of smaller galaxies and gas clouds that should produce a range of spiral galaxies rotational speeds however with little connection to galaxy luminosity. What is found however is there is a very tight correlation between rotational speed and luminosity. To explain the Tully-Fisher relationship one needs a parameter that affects both galaxy rotational speed and stellar formation (gas available and something to cause the gas to form stars) in the disk galaxy. As Disney notes there are other peculiar correlations of galaxy properties which are independent of redshift which indicates they are controlled by some factor in the galaxy rather than the environment in the vicinity of the galaxy.

There are also other interesting observational anomalies presented in the paper.

The new and old findings concerning the disk galaxy parameters and disk galaxy evolution likely should be presented in a new thread. I am interesting in this subject and have been reading other related papers concerning anomalies related to stellar formation and metallicity evolution with redshift. I will read through the papers and see if I can understand and clarify the issues. I will if I am successful start a new thread. http://arxiv.org/PS_cache/astro-ph/pdf/0610/0610031v2.pdf
A primer on hierarchical galaxy formation: the semi-analytical approach

These observations are returned to and discussed in more detail in different parts of the review.

• Why are there remarkably tight correlations between certain galaxy properties?
Spiral and elliptical galaxies exhibit tight correlations between characteristic speeds of internal motion, a structural property, and luminosity, which depends upon the star formation history (Faber & Jackson 1976; Tully & Fisher 1977; Kormendy 1977; Djorgovski & Davis 1987; Dressler et al. 1987).

• Why do we see significant changes in galaxy properties below a particular galaxy mass (Kauffmann et al. 2003)? Why are there distinct populations or a bimodality in properties such as colour (e.g. Baldry et al. 2004)?

• Why is star formation such an inefficient process? Only a small fraction of the baryons in the universe (on the order of 10%) is locked up in stars (Cole et al. 2001). Where are the remaining baryons (Fukugita, Hogan & Peebles 1998)?

• Another correlation is perhaps fundamental enough to merit its own bullet point: the correlation between the mass of the central supermassive black hole in a galaxy and the mass of the spheroidal component (Magorrian et al. 1998). Why is this relation so tight when there is such a huge difference in the spatial scale of these components? Does this correlation mean that bulges and black holes share a common formation mechanism? Did the energy released by the accretion of material onto the black hole play a role in the formation of the galaxy?

• How can we reconcile observations of seemingly massive galaxies at high redshift, some of which are forming stars at prodigious rates, with a universe in which structures grow hierarchically? What do the galaxies seen at high redshift turn? A primer on hierarchical galaxy formation into by the present day? Are we seeing the formation of today’s elliptical galaxies?

• Why is there a characteristic mass for galaxies? The most fundamental statistic describing the galaxy population is the luminosity function, a census of the number of galaxies per unit volume as a function of their luminosity. The luminosity function has a sharp break, brightwards of which the abundance of galaxies falls off exponentially (Norberg et al. 2002b; Blanton et al. 2003). The main process driving the growth of cosmic structures, gravitational instability, has no preferred scale, so processes in addition to gravity are responsible for the break.

• What role does the environment play in galaxy formation? The mix of morphological types is strongly dependent on local density, with elliptical galaxies more prevalent than spirals in the cores of clusters (Dressler 1980). The fraction of galaxies contained in groups is expected to grow with time in hierarchical models. Are the physical processes which operate within groups, such as “strangulation” or ram pressure stripping, which acts to remove the supply of cold gas in a satellite galaxy, or dynamical effects such as tidal disruption or harassment, responsible for switching off the star formation these galaxies (Gunn & Gott 1972; Moore et al. 1996; Balogh et al. 2004b; Wilman et al. 2005a; Mayer et al. 2006).
 
  • #23
Saul said:
Spiral galaxies show indicates of patterns which indicates they were not formed from many mergers.

Which patterns are these?


A second problem with the many mergers hypothesis is that simulations indicate multiple mergers of spiral with spiral produces an elliptical galaxy. Observationally that is not observed. The number of spiral galaxies stays roughly constant with redshift at 70%.

The theory that ellipticals formed from the mergers of spirals (which is by no means universally accepted) predates the theory of bottom-up structure formation and really has very little to do with it. It is also untrue that the fraction of spiral galaxies is independent of redshift -- in fact, the Hubble Sequence that we see in the local universe changes completely by z ~ 2. At z > 2, the majority of galaxies appear clumpy and irregular, suggesting heavy star formation activity and frequent mergers. I gave a talk at the Space Telescope Science Institute earlier this month on this very subject (references included):

https://webcast.stsci.edu/webcast/detail.xhtml;jsessionid=677A551BCF468F939FF85C6E2A40FFCB?talkid=1897&parent=1"



It should be noted that the paradox between simulation properties and observed properties are fundamental properties of the spiral. Angular momentum and change in angular momentum as one moves towards the core of the galaxy for example.

I'm not quite sure what this is supposed to mean. The angular momentum of stars and gas is sensitive to the galaxy's merger and star formation history, including stellar and AGN feedback. Since such things are difficult to simulate, we can't be sure that such discrepancies are resulting from a problem with the dark matter paradigm.
 
Last edited by a moderator:
  • #24
Saul said:
From a theoretical standpoint, the neutrino is required from energy/mass balance. (i.e. The point is there is a clear unequivocal theoretical requirement which indicate that the "neutrino" must exist.)

Not quite. The neutrino is only required if you demand that laws of physics obey the conservation of energy and momentum.

It does not follow from the fact there is a clear theoretical requirement that neutrinos exist and that the neutrino was found that there is a theoretical requirement for "dark matter" and that the "dark matter" exists.

Science is the proof and disproof of hypotheses. For example science move beyond phlogiston.

http://en.wikipedia.org/wiki/Phlogiston_theory

There are degrees of certainty concerning different hypotheses.

The fact that no one has been able to detect dark matter using the most sophisticated detection apparatus and that there are multiple fundamental observations which the dark matter hypothesis cannot explain, justify the criticism of the theory.

It feels like the semantic equivalent of your arguments against dark matter would be that many years of failing to detect neutrinos while at the same time finding a whole zoo of other particles that can't be explained by the neutrino hypothesis (pions, rhos, Ks, deltas, muons, and so on) is sufficient proof that neutrinos don't exist that we should abandon the idea of energy and momentum conservation.

Dark matter has specific predicted theoretical affects on rotational speeds, galaxy formation, and galaxy morphology.

As noted above those specific dark matter theoretical predictions have been shown to not be correct. In the normal process of science we would agree the theory has been falsified and the field is in a crisis. (i.e. There are multiple fundamental observations that have no explanation and the observations in question appear to be connected.)

In other fields of science the participants would not propose changing the laws of physics or to create a new particle to explain observations.

The predictions you're talking about here are contingent on the details of the physics implemented in enormous computer simulations. But, as has already been pointed out in this thread, it's a know fact that the baryonic physics in those simulations isn't good enough to capture everything going on with the baryonic matter. And, without knowing exactly how the results would be affected if those shortcomings could be fixed, you can't know whether the discrepancies with observations are shortcomings in the simulations or shortcomings in the basic physical models underlying the simulations.

One other point, eliminating dark matter would completely unravel the extremely good agreement that cosmological (rather than astrophysical) modeling has found with observational data.
 
  • #25
Parlyne said:
One other point, eliminating dark matter would completely unravel the extremely good agreement that cosmological (rather than astrophysical) modeling has found with observational data.

Good point. I think non-specialists tend to assume it's easier to model smaller things than it is bigger things. It turns out, however, that it's much easier to model the large-scale behavior of the universe than it is the formation of its component galaxies. One of the most impressive successes of the dark matter model is its ability to predict primordial element abundances (like the amount of helium relative to the amount of hydrogen) using Big Bang nucleosynthesis. The amount of dark matter relative to normal ("baryonic") matter, as predicted by nucleosynthesis models, is consistent with the amount predicted by models of the cosmic microwave background. It is difficult to get such an agreement if one assumes "dark matter" is simply parameterizing a modified theory of gravity.
 
Last edited:
  • #26
Jonathan Scott said:
As I see it, the evidence for Dark Matter consists entirely of the fact that matter on scales larger than the solar system gravitates more strongly than predicted by GR...

This is mostly true, but not entirely. The bullet cluster is a famous exception, where we see an actual offset between the distribution of dark matter (as probed by gravitational lensing) and the distribution of visible matter (as probed by broad-spectrum imaging).

http://adsabs.harvard.edu/abs/2007NuPhS.173...28C"
 
Last edited by a moderator:
  • #27
Saul said:
Jonathan,
Try to explain the observations. What you are suggesting is substituting one incorrect hypothesis with another incorrect hypothesis. Read Disney et al's paper. Think about the Tully-Fisher relationship which is a key dependent parameter but only one of the dependent parameters. There is a pattern in the spiral galaxy observations. Why?

I'm not sure what you are referring to as "another incorrect hypothesis". I personally think GR is a local approximation only, and that a better gravity theory would probably explain the success of the MOND model and at least most of the apparent need for "dark matter".

As previously mentioned on these forums, I even have one idea of how such a theory could arise, in that if the universe is closed and finite at a given time, space around large collections of masses would effectively be "conical" rather than "flat" at infinity. This gives rise to a peripheral acceleration of (c2/r) sqrt(m/M) where m/M is the effective ratio of the local collection of mass to that of the universe, and this matches MOND if M is around 1e54kg. However, this still isn't the place for pursuing that sort of speculation any further.

The well-known bullet cluster observations which SpaceTiger mentions are interesting, and certainly suggest that there's some gravitational source which we can't see, but in the absence of other supporting evidence that could of course still be theoretically explained either by a modified gravity theory or by other factors which cause the visible distribution of ordinary matter to be misleading.
 
  • #28
Saul said:
As noted in my above comments, the dark matter hypothesis does not explain spiral galaxy morphology or evolution.

I don't think it was ever intended to. If there is a reason why spiral morphology or evolution would preclude dark matter, that would be interesting, but I don't know of anything that indicates that we are at that point.

Also one of the reasons to think that dark matter exists involves observations of rotation curves, and I don't see how anything in N-body simulations invalidates that.
 
  • #29
Saul said:
What comes next if the dark matter hypothesis fails?

Modified gravity or some form of dark energy / fifth force.

Also, if you take N-body simulations and then show that they are closer to observations if you put in a modified gravity model, this would be extremely interesting. If put in a modified gravity model and show that it looks nothing at all like observations, that would be extremely interesting.

The trouble with N-body simulations is that there is enough wiggle room so it's hard to show anything conclusive and in any case you'll have to spend a year to set up the simulation.
 
  • #30
Saul said:
Dark matter appears based on observations and theoretical analysis, in published papers to be an incorrect mechanism.

The people that do simulations of galaxy-galaxy collisions as far as I know aren't confident enough with their models to make this sort of statement. There are several other independent mechanisms that favor dark matter, and no one has quite gotten any of the alternatives (modified gravity) to work quite as well.

Of course may be something that we didn't think of, but it would be interesting to try to think of something that don't fit into either dark matter or modified gravity which isn't already excluded by observations. You are welcome to try (no sarcasm here).
 
  • #31
Saul said:
If "Dark Matter" does not exist and even if dark matter did exist it could not explain the observations, at what point does astrophysics abandon the dark matter mechanism?

When some one comes up with something better.

If you can show that dark matter causes spiral galaxies calculations to go funny, this doesn't hurt dark matter at all, unless you can show that without dark matter, you get good predictions. If you get funny predictions with and without dark matter, then this suggests that the disagreement with observations has nothing to do with dark matter.

Now it would be interesting to run a N-body simulation with baryonic only matter and see what happens. One thing that I have found with computer simulations is that you have to be careful of publication bias, and that people will only publish results that are "interesting."

What would typically happens is that people will run their simulations without dark matter. If they get results that are close to observations, then it's publishable. If you run a simulation without dark matter, and the results are total non-sense then it's not publishable, since you have just confirmed what people have already suspected. What you do in these situations, is that you meet people at conferences, and they'll tell you that they did a run with something non-standard, got totally silly results, and so they didn't think this was worth publishing.

Also a lot depends on computational power. If it takes a week or two to do a simulation without dark matter, then people will do it, just as a lark to see what happens. If it takes three months to run a simulation without dark matter, then its something that really is unlikely to be done.
 
  • #32
Saul said:
Try to explain the observations. What you are suggesting is substituting one incorrect hypothesis with another incorrect hypothesis. Read Disney et al's paper. Think about the Tully-Fisher relationship which is a key dependent parameter but only one of the dependent parameters. There is a pattern in the spiral galaxy observations. Why?

I don't know.

If you have any specific ideas, I'd like to hear them. The thing about hypothesis is that they are important because once you have a hypothesis, you can start knocking them down, but without a working hypothesis, you really don't get very far.

Everyone seems to have some mental block to consider other possible explanations to what is observed and to use alternative methodologies to problem solve.

What class of possible explanations are you thinking of? The problem with possible explanations is that if you come up with a "hard target", then you usually end up with worse fits to data. Usually what happens is that it's not so much a mental block, but rather that there is some problem with the alternative explanation that renders it unviable.

Observations don't fit models. Observations *never* totally fit models, there's always something that you can't explain or that doesn't make sense. What you do is to come up with the best explanation you can.

What are the field's fundamental assumptions?

That's something really interesting to think about, and what theorists do. It would be really interesting to describe the basic assumptions that go into dark matter and modified gravity, and to figure out if there is something that doesn't fit into either category of model.

The trouble with this is that it's really hard.

Could any of the fundamental assumptions be incorrect based on the observations?

Absolutely.

What are the other observational anomalies?

They really change from month to month. Last time I had a conversation, people were interested in the fact that dwarf ellipticals didn't quite fit power spectra. I'm more of a theorist, and the two things that don't fit for me for dark matter are. There is no obvious place within the standard model for the mystery dark matter particle. Also, there is no obvious reason why galaxy distributions have to follow dark matter distributions.

I'm sure you can come up with a dozen other things that don't quite fit dark matter. Dark energy is even more speculative. However, if you come up with a dozen things that don't fit dark matter, you have to realize that there are about four big things that do fit.

The methodology people use to approach this problem is not effective. Trying to fit a round peg in a square hole for years. Part of the scientific process is identifying that there is a round peg and a square hole.

I think the methodology works fine. People will give up trying to fit a round peg into a square hole, if someone comes up with a square peg. The classic example of a situation where people went "A ha, so *that's* what's going on" is continental drift which got accepted within two years after languishing for fifty.

Trying to make a failed theory work does advance the process. It actually appears to block any progress.

A wrong hypothesis is better than no hypothesis. If there are problems, then at least you can state what the problems are.

Dark matter appears obviously to be on the ropes.

I don't think so. You've picked up one paper that has nothing obviously to do with dark matter. That's a pretty bold assumption. Dark matter is better than the obvious competitor (modified gravity). If someone comes up with another model that is neither dark matter or modified gravity, that would be interesting, and you are welcome to try.

The theory must be on the correct page.

Not true. Theories just don't match observations, because observations are always messy. When you have a mismatch, the first reaction is observational error or some small tweak that you are missing, because most of the time when you look at things further, you find that it *does* turn out to be some small tweak.

It's the game of "king of the mountain". If you want to overthrow a theory, you have to come up with something better. This is hard.
 
  • #33
SpaceTiger said:
Good point. I think non-specialists tend to assume it's easier to model smaller things than it is bigger things.

People also assume that it's easier to model "weird things" than things that you see everyday. We don't have a good model of turbulence for example, and it's far, far easier to model the entire universe and a black hole collapse than it is to model a smoke ring from a cigar.

Something that is useful to figure out complexity is the number of variables. For LCDM, there are about a dozen parameters that you can tweak. For a human being deciding what to eat for lunch, forget it.

Here is an example of something that is a physical anomaly that has not been completely explained.

http://en.wikipedia.org/wiki/Shower-curtain_effect
 
  • #34
AWA said:
Same happens with the Tooth Fairy. Rest assured though we are working hard on it.
You have been scooped by a 6-year old (I think I got the age right), Sylas' skeptical niece. Her method was inventive.
 
  • #35
There is the tooth fairy rule of theoretical astrophysics. In any theoretical astrophysics paper, you are allowed to invoke the tooth fairy once. There are some standard tooth fairies in astrophysics (turbulence, magnetic fields, atomic collective effects, various forms of observational error).

This is harder than it sounds, because often you wave the magic wand once, and you still can't explain something. The other problem (i.e. in iron core collapse supernova) is that you have about six or seven magic wands, and the trick (which no one has managed to do in about forty years of trying) is to figure out what order to wave them in.

The nice thing about dark matter, is that you wave the magic wand once, and you get about five major predictions, which makes you suspect that you are on the right track. Also by ruling out what it isn't, we've made a lot of progress in figuring out what it could be. One of two things are likely to happen. Either we narrow down the possible outcomes and then catch the animal, or we find that the constraints are fundamentally inconsistent in which case we know we messed up somewhere.
 

Similar threads

Back
Top