The Great Attractor

Why the Great Attractor matters

• CMB dipole interpretation. The ~3.36 mK temperature asymmetry in the cosmic microwave background is not cosmological — it is kinematic, reflecting our motion at ~370 km/s relative to the CMB rest frame. After subtracting Solar, Galactic, and Local Group rotation, the residual ~600 km/s peculiar flow toward Centaurus-Norma is the fingerprint of the Great Attractor's pull.
• Large-scale flow surveys. Tully-Fisher (spirals) and Fundamental-Plane (ellipticals) distance indicators give redshift-independent distances, so subtracting the Hubble flow exposes peculiar velocities. Cosmicflows-3 and Cosmicflows-4 catalogs (~56,000 galaxies) reconstruct the full velocity field out to ~600 Mly and reveal the inflow basin centered on the Great Attractor.
• Dark matter map. The infall pattern is dominated by mass, most of which is dark. Wiener filter reconstructions of the velocity field convert observed motions into a 3D dark-matter density map, with the attractor appearing as a deep gravitational well — a peak ~5σ above the cosmic mean density.
• Cosmic web visualization. Galaxies trace filaments, and the Great Attractor sits at the intersection of the Centaurus, Hydra, Pavo-Indus, and Norma filaments — a node in the cosmic web. This is the local archetype of the spider-web matter distribution that simulations like Millennium-XXL and IllustrisTNG predict on Mpc scales.
• Tests of structure formation. The amplitude of the peculiar-velocity field on 50 Mpc/h scales constrains σ₈, the matter clustering amplitude. Recent measurements give σ₈ ≈ 0.81, consistent with Planck CMB results — a non-trivial agreement between very different epochs.
• Defining the home address. Tully's 2014 watershed paper used the velocity flow to define Laniakea — our supercluster — as the gravitational basin we belong to. The Great Attractor anchors that basin. Without it, the largest "structure" we belong to would not be a structure at all.
• Future of the local universe. Comparing the infall velocity to the Hubble flow at the same distance (~5,400 km/s recession at 250 Mly) shows that even the Great Attractor cannot defeat dark energy on its scale. Knowing the boundary between bound and unbound regions tells us which neighbors will share our fate.

The numbers behind the pull

• Local Group peculiar velocity. v_pec ≈ 627 ± 22 km/s toward (l, b) = (276°, 30°), measured by the CMB dipole and confirmed by independent Tully-Fisher surveys.
• Direction. Roughly toward the Centaurus-Hydra constellations on the sky, behind Norma; in galactocentric coordinates the apex sits at (l, b) ≈ (272°, 10°) for the bulk-flow component beyond the Local Group.
• Norma Cluster mass. M₂₀₀ ≈ 1 × 10¹⁵ M_☉ from X-ray temperature (kT ≈ 8.4 keV, with a velocity dispersion of ~950 km/s); the broader Norma supercluster reaches ~3 × 10¹⁵ M_☉.
• Shapley Supercluster mass. ~10¹⁶ M_☉ distributed across roughly 25 rich clusters in a region 100 Mly across — the densest known concentration of clusters in the local universe.
• Zone of Avoidance extinction. A_V ≳ 5 mag toward Norma; in some directions A_V exceeds 30 mag, completely opaque optically. Near-infrared surveys like 2MASS reduce this to A_K ≈ 0.5 mag.
• Laniakea size. ~160 Mpc (~520 million light-years) in diameter, ~10¹⁷ M_☉ total mass, ~100,000 galaxies — defined kinematically as the watershed of the inflow velocity field.
• Hubble flow at the attractor. v_Hubble ≈ H₀ × d ≈ 70 km/s/Mpc × 77 Mpc ≈ 5,400 km/s — about 9× the infall velocity.
• Density contrast δ. The attractor region overdensity δρ/ρ̄ ≈ 1–2 averaged on 50 Mpc scales, consistent with linear-theory predictions for a 5σ peak in a Λ-CDM universe.

A brief history of the discovery

• 1976 — Smoot, Gorenstein, Muller measure the CMB dipole from a U-2 spy plane at altitude. They find a 3.5 mK temperature variation across the sky, far larger than the intrinsic CMB anisotropies later seen by COBE.
• 1986 — The "Seven Samurai" (Burstein, Davies, Dressler, Faber, Lynden-Bell, Terlevich, Wegner) publish a Tully-Fisher survey of ~400 ellipticals and find that thousands of galaxies are streaming together at ~600 km/s toward Centaurus. The streaming cannot be local; it must be driven by mass beyond.
• 1988 — Lynden-Bell and Faber name the unseen mass concentration the "Great Attractor" and estimate its mass at ~5 × 10¹⁶ M_☉, though later work pares this down once the Shapley contribution is separated.
• 1990s — ROSAT all-sky X-ray survey reveals hot intracluster gas in Norma; deep optical and HI surveys (Parkes, ATCA) map galaxies through the Zone of Avoidance.
• 2000s — 2MASS Redshift Survey and 6dFGS extend the galaxy catalog. Bulk-flow analyses pin down the inflow's amplitude and direction with percent-level precision.
• 2014 — Tully et al. publish "The Laniakea Supercluster of Galaxies" in Nature, defining our home supercluster as the watershed of inflow streamlines and locating the Great Attractor at its gravitational heart.
• Present — Cosmicflows-4 (Tully et al. 2023) reconstructs the local velocity field with 56,000 galaxy distances; the Vera Rubin Observatory and SKA precursor MeerKAT will push this catalog by an order of magnitude in the late 2020s.

Useful formulas

• CMB dipole and velocity. The fractional temperature variation across the sky is ΔT/T = (v/c) cos θ + O(v²/c²). Inverting: v ≈ c × ΔT_peak/T_CMB ≈ 3 × 10⁵ km/s × (3.36 × 10^-3 K / 2.725 K) ≈ 370 km/s. Subtracting Solar and Galactic motion gives the Local Group peculiar velocity ~627 km/s.
• Linear-theory infall. For a spherical overdensity δ on scale R, infall velocity at radius r > R is v(r) = (1/3) H₀ f(Ω_m) δ̄ r, where f(Ω_m) ≈ Ω_m^0.55. Plugging Ω_m ≈ 0.31 gives f ≈ 0.52; with δ̄ ≈ 1.5 and r ≈ 50 Mpc, v ≈ 540 km/s — the right order.
• Cluster virial mass. M₂₀₀ ≈ (5 R₂₀₀ σ²) / G, where R₂₀₀ is the radius enclosing 200× cosmic mean density and σ is line-of-sight velocity dispersion. Norma's σ ≈ 950 km/s, R₂₀₀ ≈ 2 Mpc gives M₂₀₀ ≈ 10¹⁵ M_☉.
• Hubble vs infall crossover. Bound radius r_b where v_infall(r_b) = H₀ r_b defines the boundary of gravitational binding; for Laniakea this lies near the supercluster surface, with everything outside flowing away.

Common misconceptions

• "It's a giant black hole." No. A black hole at 10¹⁶ M_☉ would have a Schwarzschild radius of ~3 × 10¹⁶ km ≈ 0.001 ly — utterly negligible at the 250 Mly scale. The Great Attractor is a diffuse gravitational basin made of dark matter, hot gas, and ~100,000 galaxies spread across hundreds of millions of light-years.
• "We'll be sucked in." No. The Hubble flow at 250 Mly is ~5,400 km/s of recession; we infall at ~600 km/s. Net motion: we drift toward it, but space grows faster. Dark energy outpaces local gravity on this scale; over cosmic time, the gap widens, not closes.
• "It's at the center of the universe." There is no center. The universe is homogeneous and isotropic on the largest scales (cosmological principle); every observer sees a similar pattern. The Great Attractor is the dominant local mass concentration but holds no privileged cosmic location.
• "It's a single object." No. It is the combined gravitational well of the Norma Cluster (~3 × 10¹⁵ M_☉), the Centaurus Cluster, several smaller groups, and — most importantly — the more distant Shapley Supercluster pulling the whole basin. The "object" is a region, not a thing.
• "We discovered it by seeing it." Backwards. We inferred its existence from CMB dipole and galaxy peculiar velocities a decade before X-ray and infrared surveys imaged the Norma Cluster through the Zone of Avoidance. Gravity told us where to look.
• "It contradicts dark matter." Opposite. The Great Attractor is among the strongest local probes of dark matter: the inflow velocity requires ~6× more mass than visible galaxies provide. The infall pattern's amplitude is one of the cleanest non-CMB tests of Λ-CDM cosmology.
• "It's the biggest structure in the universe." No. Sloan Great Wall, BOSS Great Wall, and Hercules-Corona Borealis Great Wall all exceed Laniakea's scale. The Great Attractor is impressive locally; cosmologically, it is one node among many.
• "It moves through the universe." Mostly no. The attractor itself sits at near zero peculiar velocity in the CMB rest frame on average — it is the local center of mass that everything else streams toward. The "motion" is ours, not its.

How astronomers map it

• Tully-Fisher relation. Spiral-galaxy luminosity scales with rotation speed (L ∝ v⁴). HI 21 cm line widths give v; apparent magnitude with the relation gives distance — independent of redshift. Subtracting Hubble flow yields peculiar velocity.
• Fundamental Plane. Elliptical galaxies sit on a tight 3D relation among effective radius, surface brightness, and velocity dispersion. Like Tully-Fisher but for ellipticals, used by 6dFGS to map ~9,000 nearby ellipticals.
• Type Ia supernovae. Standard-candle distances with ~5–7% precision per object; a few hundred SNe Ia within 200 Mly (e.g. ZTF, ATLAS, LSST) refine the velocity field at the largest scales.
• 21 cm HI surveys. Radio waves penetrate the Zone of Avoidance dust. Parkes HIPASS, ALFALFA, and upcoming SKA-precursor surveys catalog HI-rich galaxies behind the Galactic plane.
• X-ray cluster surveys. ROSAT All-Sky Survey, eROSITA, and Chandra detect intracluster gas (T ~ 10⁷–10⁸ K) and locate massive clusters even when optical light is extinguished — how Norma's mass was first quantified.
• Wiener filter reconstruction. Given noisy peculiar velocities at sparse points, the Wiener filter solves for the underlying density and velocity field assuming a prior power spectrum (Λ-CDM). Applied to Cosmicflows data, it produces the 3D map of the local universe published by Tully et al.