Explaining the structure of Galaxy Filaments

[kmarinas86]

Explaining the structure of Galaxy "Filaments"

Are galaxy filaments something that could be explained by the Big Bang theory, but NOT the steady state theory?

What makes the scale of galaxies so special that the forms which are generated are unlike what happens in nebulas? Is the expansion of the universe the reason for this? Why don't the movements of galaxies resemble the motions of planets in solar systems if they are both caused by the same force?

In this video it appears that the filaments themselves may just be illusions:

http://www.astro.indiana.edu/animations/b50A.mpg

Notice how if you pause the video, every now and then, you can see "structures" shift entirely every time they transition (morph).

 

[hellfire]

Both phenomena take place at very different stages of gravitational collapse. The formation of galaxy filaments and walls takes place as a transient feature after the linear regime of the collapse of an ensemble of pressureless particles in an expanding background. They evolve afterwards to non-linear three dimensional cluster structures (if matter still dominates). Stars and planetary systems form in the non-linear regime of collapse of gas clouds with pressure (jeans theory). In chapter 22 of Principles of Physical Cosmology (Peebles) there is some discussion about analytic models and simulations that lead to walls and filaments.

 

[Chronos]

This is a fractal geomety thing, in my mind. Filaments arise as a consequence of the combined effects of expansion and gravity.

 

[Wallace]

The key point is though the the structure in the Universe is not fractal, i.e. is not the same on all scales. Hellfire's post give a few good reasons why different effects at different stages of the universe's history lead to this being the case. In addition to this, different physics occurs on different scales, an object like a star is not stable beyond a certain size due to the inherent nuclear physics and thermodynamics of the material, so you couldn't not have a galaxy sized 'star'.

In a similar way there is gas physics that influences the structure of galaxies that does not have an effect on the scale of galaxy clusters.

 

[Chronos]

Depends on how you define fractal structures. It is obvious structure formation is not the same at all scales. I no more expect to see a galaxy sized star than a stellar sized atom. Chaos theory [fractal patterns] is not that simplistic.

 

[kmarinas86]

Maybe galaxy filaments are illusions

In this video it appears that the filaments themselves may just be illusions:

http://www.astro.indiana.edu/animations/b50A.mpg

Notice how if you pause the video, every now and then, you can see "structures" shift entirely every time they transition (morph).

It appears that the real galaxial cluster is much larger than the apparent cluster. In fact, it appears that this video shows a galaxial supercluster that is actually formed by many smaller psuedo clusters which are illusions by the proximity of galaxies passing by. Granted some of the galaxies do not move as much, but it seems that significant changes in the structure are due to a sort of "marching band" effect rather than gravitational cohesion.

Am I just seeing things, or is this possible with current astronomical evidence?

This video makes it very convincing:
http://www.astro.indiana.edu/animations/b50A.mpg

The velocities are high enough for the majority of orbital trajectories to be extremely elliptical. These elliptical trajactories have orbital motions which have larger radius than radius of their immediate cluster. This begs the question of dark matter. Granted, if it were known that the centers of clusters are going at closer to their expected speeds than their "suburban galactic" counterparts, we might think that there is gravitational cohesion of galaxies, but only when they are at very close proximity. This is especially true if we think of dark matter halo as being more concentrated at the "suburban and rural" areas of the "apparent" galaxy clusters where there is a greater difference between velocity expected from visible mass and the velocity expected from virial mass.

 

[Allday]

Some thoughts

Filamentary structure in the universe is a research topic that is very much in flux right now. The trickiest part is actually coming up with a definition of "filament" or "pancake" or "wall". There are several recent (2006) papers on the astro-ph arxiv that deal with this subject.

Large N-body simulations can be used to try to quantify large scale structure. Clusters in these simulations can be detected and classified by their moment of inertia tensors. If one principal axis is much longer than the other two then you can label that cluster as a filament. Of course this method has its drawbacks. There are other methods like using a minimal spanning tree (http://arxiv.org/abs/astro-ph/0611641) , or using an algorithm to find the voids in the simulation and then defining the filamments and sheets at the intersections of these voids.

The point I am trying to make is that one can identify "by eye" the filaments and walls in a simulation with the proper visualization. (usually this means looking at the dark matter) but its much harder to quantify large scale structure. By the way, it seemed that the visualization you linked to was of the luminous matter (but I am just guessing) Also the scale that filaments exist on is 10's - 100's of Mpc. Galaxy clusters form at the intersection of filaments. Filaments themselves can contain many milky way sized galaxies.

I could go on about the observational side if anyone is interested ...

 

[hellfire]

I could go on about the observational side if anyone is interested ...

Yes, please. Feel free to elaborate. I would be especially interested in understanding the claim that filaments are largely dominant and walls are made of a set of connected filaments. Does this claim have observational support?

 

[Allday]

Its very tricky to make a connection between observation and simulation in this area. The major reason is that the large scale structure is dominated by dark matter and observations are dominated by luminous matter.

The large scale surveys like SDSS have compiled galaxy catalogues where a "cosmic web" can be seen, but the luminous galaxies are just tracers of the dark matter which dominates large scale structure. A very big unknown is the bias that is introduced here. Some people use semi analytic models to populate the dark matter halos identified in simulations with luminous galaxies BUT these models are just good guesses.

One way to make a slightly better connection is to use the diffuse intergalactic medium as a tracer of the dark matter instead of the luminous matter. This can be done using high redshift quasar spectra. The lyman alpha forrest in these spectra is a record of the neutral hydrogen density along the line of sight to the quasar. These can be statistically compared to mock spectra made by ray tracing through the simulation box and assuming a dark matter/normal matter ratio. The filamentary structure in simulations leads to a good agreement between the mock spectra and real spectra, but assumptions have to be made here as well. Also ionized hydrogen will not leave its mark on the spectra and so it is not a complete tracer of the dark matter

I have worked on identifying an alignment signal between filaments and galaxy sized dark matter halos (ie dark matter halos that should host milky way sized galaxies) and there is a signal in the simulations, but it would be difficult to detect observationally. Also I have heard of some work on an obsservational alignment signal between spiral galaxies and void walls, but I can't find the link right now.

As far as the claim that filaments dominate and that sheets are made of groups of filaments, i think the observational evidence for that will have to come after someone figures out a good way to actually quantitatively observe filaments.

hope that was clear, ask me if anything need shoring up

 

[hellfire]

Also I have heard of some work on an obsservational alignment signal between spiral galaxies and void walls, but I can't find the link right now.

Do you mean something like this:

The orientation of galaxy dark matter haloes around cosmic voids
http://arxiv.org/abs/astro-ph/0609629

I read this paper some months ago and found it rather difficult to imagine or picture out what a mechanism can lead to such alignments.

They make some interesting claims:

The alignment of the haloes with their local large–scale structure is not only of interest to constrain models of galaxy formation, it could also be relevant to explain other observational features. For example, the tendency of satellite galaxies to avoid orbits that are coplanar with their host spiral galaxies (know as the ”Holmberg effect”) found in observations (Holmberg 1969; Zaritsky et al. 1997; Sales & Lambas 2004, but see Agustsson & Brainerd 2006) and in simulations (Zentner et al. 2005; Libeskind et al. 2006). This could be due to the preferential accretion of satellites along filaments, that we have seen are preferentially aligned with the major axis of the host halo.

 

[Allday]

Hi hellfire, sorry for the late response. I had been referring to that paper and the observation they mention in the abstract. I find it difficult to imagine a physical mechanism for this alignment other than the void walls are where the filaments and sheets must be located. It is an area i haven't kept up on as much as i used to. I am trying to write a radiative transfer code and that's been taking up the majority of my time.

-best
gabe

 

[heusdens]

In this video it appears that the filaments themselves may just be illusions:

http://www.astro.indiana.edu/animations/b50A.mpg

Notice how if you pause the video, every now and then, you can see "structures" shift entirely every time they transition (morph).

It appears that the real galaxial cluster is much larger than the apparent cluster. In fact, it appears that this video shows a galaxial supercluster that is actually formed by many smaller psuedo clusters which are illusions by the proximity of galaxies passing by. Granted some of the galaxies do not move as much, but it seems that significant changes in the structure are due to a sort of "marching band" effect rather than gravitational cohesion.

Am I just seeing things, or is this possible with current astronomical evidence?

This video makes it very convincing:
http://www.astro.indiana.edu/animations/b50A.mpg

The velocities are high enough for the majority of orbital trajectories to be extremely elliptical. These elliptical trajactories have orbital motions which have larger radius than radius of their immediate cluster. This begs the question of dark matter. Granted, if it were known that the centers of clusters are going at closer to their expected speeds than their "suburban galactic" counterparts, we might think that there is gravitational cohesion of galaxies, but only when they are at very close proximity. This is especially true if we think of dark matter halo as being more concentrated at the "suburban and rural" areas of the "apparent" galaxy clusters where there is a greater difference between velocity expected from visible mass and the velocity expected from virial mass.

I don't know how that simulation was made, but it seems that it was made doing a rather small portion of the univere. As the universe stretchess around much larger, the combined forces of the rest of the universe might have simply been ignored, why you probably see much more and/or faster movements then when that was calculated.

But doing that calculation is rather difficult, you would have to size up the simulation to very large size (while showing only a small portion for it), to compensate for that.