Tuesday, December 3, 2013

Local Group Astronomy Data Favors Warm Dark Matter Over Cold Dark Matter

As an aside, I'll note here that detailed astronomy data from the Milky Way and our Local Group (LG) of galaxies has been used to infer fairly precisely the distribution of dark matter in our LG, which the paper then uses to establish a variety of constraints on the properties of dark matter generally by looking at which assumptions fed into a computer model (called CLUES) can reproduce the observed reality.

On balance, this analysis clearly favors "warm dark matter" (WDM) models (which typically assume dark matter particles with masses on the order of low keV each) over "cold dark matter" (CDM) models (which typically assume dark matter particles with masses on the order of 10s to 100s of GeV each). There is one respect in which the LG data slightly favors CDM over WDM, but not to the extent that it excludes a WDM model. The conclusion of the papers summarizes that analysis as follows:
Using same initial conditions, numerical resolution and sub-grid physics models the CDM and WDM simulations were performed and compared. . . . Apart from the expected di fferences with respect to the abundance and distribution of satellites, and their associated baryonic physics [Ed. which strongly favors WDM models], a new interesting result has been found. 
The two simulated LG-like objects are in diff erent stages of evolution. The CDM LG is beyond its turn-around phase and is more compact. The WDM object, on the other hand is dynamically younger, more diff use and has not reached turned-around. The interesting aspect of the di fference between the two cases is that in spite of the fact that the CDM and WDM power spectra coincide on the scale of the LG, namely for mass scale larger than 1012h-1Mpc, they do diff er dynamically. It follows that the cross-talk between diff erent scales a ffects the dynamics on the LG scale. This does not mean, of course, that a proper LG-like object cannot be found in a WDM scenario. But, compared with CDM, it would be less likely to fi nd such an object in an environment constrained to mimic the one in which the LG formed.

The other approach to using the Local Universe as a DM laboratory is by using full box DM-only constrained simulations, assume some simplifi ed model for populating DM halos with galaxies, and then compare the outcome with local surveys of galaxies. By varying the nature of the DM, and hence the power spectrum of the initial condition, while keeping all other aspects of the simulations intact, one can set stringent constraints on the nature of the DM or the model used to associate galaxies with halos. This is the approach adopted by Zavala et al. (2009)) and by Tikhonov et al. (2009). Using, what might be considered a very naive model of associating galaxies with DM halos, the comparison with the observed velocity function of the ALFALFA galaxies (Giovanelli et al., 2005) and the spectrum of mini-voids in the very Local Universe (Tikhonov and Klypin, 2009) clearly favours the WDM model on the CDM one.
Implications For Future Dark Matter Research

This has implications for what kind of experimental searches for dark matter make sense to conduct in the future.

* The LG analysis adds to a long litany of astronomical observations that prefer WDM over CDM, and it has already been established that the LamdaCDM model applied to measurements of the Cosmic Microwave Background (CMB) like WMAP and Planck is actually indifferent to the distinction between WDM and CDM models.

* Similar studies analyzing astronomy data and comparing them to simulations with various assumptions about dark matter properties also disfavor dark matter with more than one predominant type of dark matter particle and self-interacting dark matter models.

* LUX has largely ruled out dark matter with the abundance that should be present in the vicinity of our solar system, that has a cross-section of interaction as large as a neutrino for masses of about 5 GeV to almost 1000 GeV, which includes essentially all "weakly interacting massive particles" in the technical sense of particles that have weak force interactions with W and Z bosons.

* Cold dark matter heavier than 1 TeV should be even more strongly disfavored by the astronomy observations that favor WDM over CDM than WIMPS in the 100 GeV range (which are already strongly disfavored).

* Precision electroweak measurements have also pretty definitively established that there are no weakly interacting neutral particles that would create "invisible decays" with masses of less than one-half of the Z boson mass (about 45 GeV) other than exactly three kinds of neutrinos.  Even a fractional weak force coupling would have to be less than, or on the order of 0.2% of the strength of a neutrino coupling (i.e. one five hundredth) to fit the precision electroweak data and there is absolute no precedent in Standard Model physics for any such tiny fractional weak force coupling.  W and Z boson decays are "democratic" and come in neat integer proportions.

In short, the data increasingly strongly point to a dark matter particle, if there is one, that is on the order of a 1-3 keV in mass, with a characteristic average speed, that do not interact via the weak nuclear force, the electromagnetic force or the strong force, with a quite well defined abundance in the vicinity of the solar system.

Direct dark matter detection experiments that are looking for anything else are a priori probably a waste of money with respect to pinning down the nature of dark matter and not well motivated theoretically at this point, although they may discover interesting things about neutrinos and developing them may promote a better understanding of the background flux of familiar particles of all kinds that a detector would have to rule out in the vicinity of Earth.

No direct dark matter experiment is capable of detecting such a light particle, particularly when it is entirely collisionless and interacts solely via gravity except to the extent that there are Fermi contract interactions in which a dark matter particle heads towards exactly the same place in time and space as another Fermion which given its character as a fermion, it cannot.  But, this would require two objects with characteristic lengths for this purpose each of the order of the tiny Planck length and even then the tiny mass of the keV mass warm dark matter particles that aren't moving at relativistic speeds would not be immense relative to the mass and momentum of a typical electron or quark (contract with a neutrino might be virtually undetectable anyway).

Similarly, there is no well motivated reason to believe that the LHC is likely to discover a new keV mass electromagnetically neutral particle without QCD color charge or electromagnetic charge as its data set gets larger and it moves to higher energies than it has already probed.  The one exception to that would be possible anomalous decays or other non-Standard Model properties of the Higgs boson, on the theory that dark matter particles might interact with the Higgs field even though it they do not interact with the weak force.  But, even this would be something that would have to be discovered in future Higgs factories.  The LHC itself will probably never have the precision necessary to pick up that kind of very subtle signal with meaningful statistical significance.

Finally, while dark matter might annihilate in some fashion, we have no good models that fit the preferred parameter space of dark matter that call for this to happen.  Moreover, if dark matter did annihilate by some means, it should produce quite diffuse and low energy photons and cosmic rays (which despite the term often refers to fast moving cosmic fermions) that would be difficult or impossible to distinguish from background events that we believe we understand (although only dimly).  High energy cosmic rays at energies approximating those found at the Fermi line are poor candidates for dark matter annihilation signatures.

Warm dark matter, if it exists, may very well be so ephemeral that it can only be detected indirectly via gravitational interactions, and may thus, for all practical purposes for the foreseeable future be impossible to detect directly.  Astronomy data is realistically the only way to better understand dark matter, if it exists, going forward.

In principle, at least, it should be easier to directly search for modified gravity effects with deep space probes in weak gravitational fields, than to detect warm dark matter directly.  Such experiments ought to be able to distinguish the two.


Anonymous said...

Great post.
I wish that the Arxiv article you reference did a better job of creating a plot that showed (a) CDM (b) WDM, and (c) experimental data.
I couldn't find a plot that actually had experimental data on the same plot as simulation data.
Am I mistaken?

Also, I thought that you'd be interested to know that the writers at New Scientist are still in denial about keV darm matter. Did you see their article on 0.11meV axion dark matter?

Dark matter less than an eV has been ruled out for decades. I thought that the quality of New Scientist articles couldn't fall any lower. Now they've just gone and proved me wrong.

andrew said...

FWIW, even less than rigorous popularization of science magazines like New Scientist embrace fewer loony ideas than the collective contributor class to arxiv. I don't read the New Scientist, however.

I probably ought to do a blog post exploring the dark matter mass-medium scale structure connection. There are assumptions that go into that relationship and if one of those assumptions is wrong than lighter dark matter would be more plausible. The speed of the particle's inferred from their mass is really more important than the mass itself, and the speed-mass relationship really only applies to "relic" dark matter generated early in the universe before it freezes out. Hot dark matter would also create an Neff signal, however, unless, again, it does not have a "relic" source.

I don't know if axion dark matter theorists have found a way to address these issues. If dark matter isn't a relic of the early universe for CMB cosmology purposes, however, the lamdaCDM model shouldn't work, and it does, so the assumptions used to rule out hot dark matter are probably pretty sound.