The evidence for dark matter

Dark matter is not a hypothesis reached for lightly — it is forced on us by independent measurements spanning galaxies, clusters, gravitational lensing, and the cosmic microwave background, all of which agree that most of the matter neither shines nor is baryonic.

Galaxy rotation curves

The cleanest evidence is kinematic. If a galaxy's mass followed its light, the circular speed beyond the luminous disk would fall as $v\propto r^{-1/2}$ (Keplerian). Instead, measured rotation curves stay flat out to large radii (figure) — which requires the enclosed mass to keep growing, $M(

Worked example — the mass a flat curve demands

For a circular orbit, $v^2=GM( $$M(<20\,{\rm kpc})=\frac{(2\times10^{5})^2(6.2\times10^{20})}{6.67\times10^{-11}}\approx3.7\times10^{41}\ {\rm kg}\approx1.9\times10^{11}\,M_\odot,$$

several times the stellar mass — and it keeps rising with $r$. The missing mass is dark.

A galaxy rotation curve: the observed speed stays flat (blue) where visible matter alone predicts a Keplerian falloff (dashed). The shaded gap is the mass contributed by the dark halo.

Clusters and the virial discrepancy

Zwicky (1933) applied the virial theorem to the Coma cluster and found the galaxies move far too fast to be bound by their visible mass — the first hint of dark matter, on the largest scales. Modern cluster dynamics and X-ray gas temperatures confirm a large dark component.

Gravitational lensing

Light bends around mass regardless of whether it shines. Weak and strong lensing map cluster mass directly, and in the Bullet Cluster the lensing mass is spatially offset from the X-ray gas (most of the baryons) — showing the dominant mass is collisionless and dark, not the gas.

The CMB and large-scale structure

The most precise measurement is cosmological. The heights of the acoustic peaks in the CMB power spectrum require about five times more matter than baryons — a non-baryonic component that does not couple to photons. The same $\Omega_c$ then correctly predicts the growth of large-scale structure. Independent probes converge on $\Omega_c\approx0.26$.

In our research

That $\sim$5:1 dark-to-baryon ratio is the $\Omega_c\approx0.26$ we put into every simulation. What none of this evidence tells us is what the dark matter is — cold particles or an ultralight wave. Distinguishing those on the galactic scales where they differ is the entire point of this program.

Key references

Rubin & Ford (1970), Rotation of the Andromeda Nebula, ApJ 159, 379.
Clowe et al. (2006), A direct empirical proof of dark matter (Bullet Cluster), ApJ 648, L109.
Planck Collaboration (2020), A&A 641, A6.

PreviousCosmic epochs relevant to our work (z=127→0)NextCold dark matter and its small-scale problems