Dark matter and dark energy

Roughly 95% of the energy content of the universe is in two components that have never been directly detected. They are unrelated to each other, and the evidence for each is strong but entirely gravitational.

Evidence for dark matter

The case does not rest on any single observation. It rests on several independent measurements at different scales that require the same amount of extra mass.

The Bullet Cluster is the observation that most directly resists alternatives. Two galaxy clusters passed through each other. The hot gas, which is most of the baryonic mass, interacted and was slowed, pooling between the clusters. The galaxies, effectively collisionless, continued. Weak lensing shows the gravitational mass followed the galaxies, not the gas.

If gravity were simply modified, mass would track the dominant baryonic component, which is the gas. It does not. The lensing mass is where the collisionless component is, which is what a collisionless dark matter halo predicts.

The CMB evidence is independent and equally strong. The relative heights of the acoustic peaks depend on how much matter participates in photon-baryon oscillations against how much only gravitates. Fitting them requires roughly five times more non-baryonic than baryonic matter, matching the value from completely different methods. See Cosmology.

Candidates

WIMPs were attractive because a particle with weak-scale mass and weak-scale interaction cross-section naturally produces the observed abundance, a coincidence known as the “WIMP miracle”, and because supersymmetry independently predicted such particles.

Direct detection experiments look for rare nuclear recoils from dark matter particles passing through ultra-sensitive detectors deep underground. XENONnT, LZ and PandaX have pushed exclusion limits down by orders of magnitude over two decades without a detection.

These experiments are approaching the neutrino fog, the point at which coherent scattering of astrophysical neutrinos produces an irreducible background resembling the signal. Beyond it, improving sensitivity by making the detector larger stops working. That is a real horizon for the technique rather than a temporary obstacle, and it has shifted attention toward axions and other candidates.

Modified gravity

An alternative is that dark matter does not exist and gravity behaves differently at low acceleration. MOND proposes a modification below a characteristic acceleration scale of about $10^{-10}$ m/s².

The honest assessment has two halves.

MOND is remarkably successful at galaxy scales. It reproduces flat rotation curves and predicts the Tully-Fisher relation as a consequence rather than fitting it, using one parameter. Some galaxy-scale regularities it predicts were observed afterwards. A theory that predicts is doing something a curve fit is not.

MOND fails at larger scales. It does not account for cluster dynamics without adding unseen mass anyway, does not explain the Bullet Cluster separation, and does not reproduce the CMB acoustic peak structure. Relativistic extensions exist but are contrived and were further constrained by GW170817, which showed gravitational waves travel at the speed of light to within one part in 10¹⁵, ruling out several formulations. See Multi-messenger astronomy.

The current consensus is that dark matter is correct and that MOND’s galaxy-scale successes represent a real regularity that dark matter models should explain better than they currently do.

Dark energy

Dark energy is the larger mystery and the more recent one.

In 1998 two teams independently found, using Type Ia supernovae as standardisable candles, that distant supernovae were fainter than expected. The expansion of the universe is accelerating.

The simplest explanation is a cosmological constant $\Lambda$: a constant energy density of the vacuum with negative pressure. It fits essentially all data.

The difficulty is theoretical rather than observational. Quantum field theory predicts a vacuum energy density, and the predicted value exceeds the observed one by roughly 120 orders of magnitude. This is often called the worst quantitative prediction in physics, and no accepted resolution exists.

Cosmological constant

Fits the data with one parameter. The equation of state w = -1 is consistent with all current measurements.

Quintessence

A dynamical field whose energy density evolves, so w differs from -1 and may change with time. Would be distinguishable with better data.

Modified gravity at cosmological scales

Acceleration as a departure from general relativity rather than a new component. Increasingly constrained.

The coincidence problem

Dark energy has only recently come to dominate. That we live at the moment when matter and dark energy densities are comparable requires explanation or is a selection effect.

Current and forthcoming surveys, DESI, Euclid, Rubin and Roman, are designed to measure the equation of state parameter $w$ and any evolution in it. Early DESI results have hinted at evolution, which if confirmed would rule out a plain cosmological constant and be the most significant cosmological result in decades. The evidence is not yet conclusive.

References

• Bertone, G. and Hooper, D. “History of dark matter”, Reviews of Modern Physics, 90:045002, 2018.
• Clowe, D. et al. “A Direct Empirical Proof of the Existence of Dark Matter”, The Astrophysical Journal Letters, 648:L109, 2006. The Bullet Cluster.
• Riess, A. G. et al. “Observational Evidence from Supernovae for an Accelerating Universe”, The Astronomical Journal, 116:1009, 1998, and Perlmutter, S. et al., ApJ, 517:565, 1999.
• Famaey, B. and McGaugh, S. S. “Modified Newtonian Dynamics: A Review”, Living Reviews in Relativity, 15:10, 2012.
• Weinberg, S. “The cosmological constant problem”, Reviews of Modern Physics, 61:1, 1989.
• Billard, J. et al. “Implication of neutrino backgrounds on the reach of next generation dark matter direct detection experiments”, Physical Review D, 89:023524, 2014.