The distribution of matter on scales of megaparsecs to gigaparsecs, encompassing galaxy clusters, superclusters, filaments, walls, and voids — collectively known as the cosmic web.

## Key Points

- **Walls/sheets**: planar overdense structures; galaxies distributed in thin layers (e.g., CfA Great Wall, Sloan Great Wall at ~420 Mpc long); collapse along one axis
- **Nodes/knots**: intersections of filaments where galaxy clusters reside; highest density environments; contain ~5-10% of mass
- **Voids**: underdense regions occupying ~60-80% of the universe's volume; typical diameter 10-100 Mpc; nearly empty but contain sparse galaxies; density ~10-20% of mean
- Total mass: 10^14 to 10^15 solar masses
- Virial radius: 1-3 Mpc
- Velocity dispersion: 500-1500 km/s
- ICM temperature: 2-15 keV (2 x 10^7 to 1.7 x 10^8 K)
- X-ray luminosity: 10^43 to 10^46 erg/s
- Number density: ~10^-5 to 10^-6 per Mpc^3 for massive clusters (>10^14.5 solar masses)
- **Virgo Cluster**: nearest major cluster; ~16.5 Mpc; ~1300 galaxies; centered on M87; irregular, still assembling; M87 jet (first imaged BH by EHT)
- **Coma Cluster**: prototype rich cluster; ~100 Mpc; first evidence for dark matter (Zwicky, 1933); ~1000 large galaxies; regular, relaxed morphology; velocity dispersion ~1000 km/s
- **Perseus Cluster** (Abell 426): brightest X-ray cluster; ~73 Mpc; sound waves in ICM (Chandra); NGC 1275 central AGN; cooling flow cluster

## Quick Example

```
d^2(delta)/dt^2 + 2H * d(delta)/dt = 4*pi*G*rho_mean * delta
```

```
lambda_J = c_s * sqrt(pi / (G * rho))
```

Large-Scale Structure of the Universe

The distribution of matter on scales of megaparsecs to gigaparsecs, encompassing galaxy clusters, superclusters, filaments, walls, and voids — collectively known as the cosmic web.

---

The Cosmic Web

The universe's matter distribution on scales >1 Mpc forms a network-like pattern:

• Filaments: elongated overdense regions containing chains of galaxies; typically 10-100 Mpc long, few Mpc wide; contain ~40-50% of all baryonic matter; arise naturally from gravitational collapse of initial density perturbations along two axes
• Walls/sheets: planar overdense structures; galaxies distributed in thin layers (e.g., CfA Great Wall, Sloan Great Wall at ~420 Mpc long); collapse along one axis
• Nodes/knots: intersections of filaments where galaxy clusters reside; highest density environments; contain ~5-10% of mass
• Voids: underdense regions occupying ~60-80% of the universe's volume; typical diameter 10-100 Mpc; nearly empty but contain sparse galaxies; density ~10-20% of mean

This structure emerges naturally from gravitational amplification of small initial density fluctuations (delta-rho/rho ~ 10^-5 at recombination) imprinted during inflation.

---

Galaxy Clusters

Galaxy clusters are the most massive gravitationally bound objects in the universe.

Composition

Component	Mass Fraction	Description
Dark matter	~80-85%	Dominates gravitational potential; detected via lensing and dynamics
Intracluster medium (ICM)	~12-15%	Hot diffuse gas (10^7 - 10^8 K); emits X-rays via thermal bremsstrahlung; contains most baryonic mass
Galaxies	~3-5%	Hundreds to thousands of galaxies; dominated by central massive elliptical (BCG)

Key Properties

• Total mass: 10^14 to 10^15 solar masses
• Virial radius: 1-3 Mpc
• Velocity dispersion: 500-1500 km/s
• ICM temperature: 2-15 keV (2 x 10^7 to 1.7 x 10^8 K)
• X-ray luminosity: 10^43 to 10^46 erg/s
• Number density: ~10^-5 to 10^-6 per Mpc^3 for massive clusters (>10^14.5 solar masses)

Notable Clusters

• Virgo Cluster: nearest major cluster; ~16.5 Mpc; ~1300 galaxies; centered on M87; irregular, still assembling; M87 jet (first imaged BH by EHT)
• Coma Cluster: prototype rich cluster; ~100 Mpc; first evidence for dark matter (Zwicky, 1933); ~1000 large galaxies; regular, relaxed morphology; velocity dispersion ~1000 km/s
• Perseus Cluster (Abell 426): brightest X-ray cluster; ~73 Mpc; sound waves in ICM (Chandra); NGC 1275 central AGN; cooling flow cluster
• Bullet Cluster (1E 0657-56): two sub-clusters post-collision; X-ray gas separated from lensing mass; key evidence for dark matter; collision velocity ~4700 km/s

Physics of the ICM

• Cooling flows: radiative cooling time in cluster centers < Hubble time; gas should cool and flow inward at 100-1000 solar masses/yr; actual cooling rates 10x lower than predicted (the "cooling flow problem")
• AGN feedback: central AGN injects energy via jets and bubbles that heat ICM, offsetting cooling; observed as X-ray cavities (Perseus, MS 0735); self-regulating cycle
• Sunyaev-Zel'dovich (SZ) effect: CMB photons inverse-Compton scatter off hot ICM electrons; spectral distortion independent of redshift — powerful for finding distant clusters; thermal SZ (y-parameter) and kinetic SZ (peculiar velocity)
• Metal enrichment: ICM metallicity ~0.3-0.5 solar; iron from Type Ia supernovae in cluster galaxies; oxygen/silicon from core-collapse supernovae

Cluster Mass Estimation Methods

1. Galaxy dynamics: virial theorem M ~ R*sigma^2/G; subject to velocity anisotropy assumptions
2. X-ray hydrostatic: assume ICM in hydrostatic equilibrium; M(r) from temperature and density profiles; ~10-20% bias from non-thermal pressure
3. Gravitational lensing: strong lensing (arcs near core) + weak lensing (statistical shear at large radii); direct mass measurement independent of dynamical state
4. SZ effect: integrated SZ signal Y correlates with total mass
5. Richness/optical: number of member galaxies correlates with mass (with scatter); used for large surveys

---

Galaxy Groups

• Most common galaxy environment; 3-50 members; mass 10^12.5 to 10^14 solar masses
• Local Group: ~80+ known members; dominated by Milky Way (~10^12 solar masses) and Andromeda (M31, ~1.2 x 10^12 solar masses); Triangulum (M33) third largest; total mass ~2-3 x 10^12 solar masses; diameter ~3 Mpc
• Fornax Cluster/Group: ~60 Mpc; rich concentration of early-type galaxies; NGC 1399 central elliptical; important for studying environmental quenching
• Compact groups (Hickson): 4-8 galaxies in close proximity; high interaction rate; rapid evolution; may merge into single elliptical; HCG 92 (Stephan's Quintet, JWST first image)
• Groups contain majority of galaxies in the universe; critical environment for galaxy preprocessing before cluster infall

---

Superclusters

Superclusters are the largest coherent structures, though generally not gravitationally bound (expansion will eventually disperse them).

Notable Superclusters

• Laniakea: our home supercluster; identified by Tully et al. (2014) using velocity flow mapping; ~160 Mpc diameter; ~10^17 solar masses; ~100,000 galaxies; includes Virgo, Centaurus, Hydra clusters; basin of attraction centered near the Great Attractor
• Shapley Supercluster (SCl 124): most massive concentration in the nearby universe (~200 Mpc); ~8000 galaxies in central region; multiple merging clusters; significant contributor to the Local Group's peculiar velocity
• Great Attractor: gravitational anomaly at ~65 Mpc toward Centaurus-Norma; originally identified from CMB dipole residuals and galaxy peculiar velocities; lies behind the Zone of Avoidance (galactic plane obscuration); associated with Norma Cluster (Abell 3627)
• Sloan Great Wall: ~420 Mpc long; one of largest known structures; discovered in SDSS data (2003)
• Hercules-Corona Borealis Great Wall: ~3 Gpc; identified from gamma-ray burst distribution; status debated (may violate homogeneity scale)

---

Baryon Acoustic Oscillations (BAO)

Physical Origin

Before recombination (z > 1100), photon-baryon fluid supported acoustic (sound) waves:

• Overdense regions drove expansion against gravity; photon pressure resisted compression
• At recombination, photons decoupled; acoustic waves froze in place
• The sound horizon at decombination: r_d ~ 147 Mpc (comoving) — this is the distance sound traveled from the Big Bang to recombination

Standard Ruler

The sound horizon imprints a characteristic scale in:

• CMB: first acoustic peak at l ~ 220 (angular scale ~ 1 degree corresponds to ~147 Mpc at z ~ 1100)
• Galaxy distribution: excess probability of finding galaxy pairs separated by ~147 Mpc (comoving); first detected in SDSS by Eisenstein et al. (2005)

BAO provides a standard ruler for measuring:

• D_A(z): angular diameter distance (transverse BAO; from angular separation of galaxy pairs)
• D_H(z) = c/H(z): Hubble distance (radial BAO; from redshift-space separation)
• Combined: D_V(z) = [z * D_A(z)^2 * D_H(z)]^(1/3) (volume-averaged distance)

DESI BAO Results (2024-2025)

The Dark Energy Spectroscopic Instrument measured BAO in 7 redshift bins using >5.7 million tracers:

Tracer	Redshift Range	Effective z
Bright Galaxy Survey (BGS)	0.1-0.4	0.30
Luminous Red Galaxies (LRG1)	0.4-0.6	0.51
LRG2	0.6-0.8	0.71
LRG3 + ELG1	0.8-1.1	0.93
Emission Line Galaxies (ELG2)	1.1-1.6	1.32
Quasars (QSO)	0.8-2.1	1.49
Lyman-alpha forest	1.77-4.16	2.33

Key findings: combined with CMB and Type Ia supernovae, DESI data show >2 sigma preference for time-varying dark energy equation of state over cosmological constant. The w_0-w_a parameterization suggests w evolved from w > -1 in the past toward w < -1 recently.

---

Cosmic Voids

Properties

• Typical diameter: 10-100 Mpc; largest voids >100 Mpc (Bootes void ~110 Mpc diameter)
• Occupy ~60-80% of universe's volume but contain <20% of its mass
• Interior density: ~10-20% of cosmic mean
• Not completely empty: contain sparse galaxy populations (void galaxies tend to be bluer, more gas-rich)
• Void shapes: mildly oblate in real space; appear elongated along line of sight in redshift space (Alcock-Paczynski effect)

Cosmological Applications

• Integrated Sachs-Wolfe (ISW) effect: photons traversing voids in a dark-energy-dominated universe gain net energy (blueshift > redshift because void grows during traversal); detected as CMB cold spots correlated with void positions
• Alcock-Paczynski test: void shapes in redshift space depend on cosmological parameters; spherical voids become distorted if wrong cosmology assumed
• Modified gravity tests: void interiors are low-density environments where fifth-force effects (chameleon, f(R)) are less screened; void profiles differ between GR and modified gravity
• Void lensing: voids produce detectable (negative) weak lensing signal; constrains void density profiles and cosmology

---

Structure Formation

Gravitational Instability

Linear perturbation theory (for delta = delta-rho/rho << 1):

d^2(delta)/dt^2 + 2H * d(delta)/dt = 4*pi*G*rho_mean * delta

• Growing mode: delta(t) proportional to D(t), the linear growth factor
• In matter domination: D proportional to a (scale factor)
• Dark energy slows growth: structures grow more slowly at late times

Jeans Length

Minimum scale for gravitational collapse against pressure:

lambda_J = c_s * sqrt(pi / (G * rho))

• For baryons before recombination: lambda_J ~ 100 Mpc (coupled to photons); too large for structure to grow
• After recombination: lambda_J drops dramatically (~kpc scale); baryons fall into pre-existing dark matter potential wells
• Dark matter has no pressure support; begins growing structure immediately after matter-radiation equality (z ~ 3400)

Dark Matter Scaffolding

Dark matter structures form first and baryons follow:

1. Dark matter perturbations grow from z ~ 3400 onward (not impeded by photon pressure)
2. Smallest halos form first; merge hierarchically into larger structures (bottom-up)
3. After recombination, baryons fall into dark matter potential wells
4. This explains why galaxy clustering traces dark matter distribution with a bias factor b: delta_gal = b * delta_DM

N-body Simulations

Simulation	Year	Particles	Box Size	Key Contribution
Millennium	2005	10^10	500 Mpc/h	First large-volume simulation with semi-analytic galaxy models; established halo merger trees
Millennium-XXL	2012	6.7 x 10^10	3 Gpc/h	Cluster and supercluster statistics
Illustris	2014	~10^10	106 Mpc	First full magneto-hydrodynamic cosmological simulation with galaxy formation physics
IllustrisTNG	2018	up to 3 x 10^10	50-300 Mpc	Improved feedback models; reproduced galaxy color bimodality, clustering, quenched fractions
FLAMINGO	2023	up to 3 x 10^11	2.8 Gpc	Largest hydro simulation; calibrated to observed gas fractions; cluster counts and S_8
AbacusSummit	2021	6.9 x 10^13 total	multiple	Suite of 150+ N-body sims for DESI analysis; high-accuracy halo statistics
Uchuu	2021	2.1 x 10^12	2 Gpc/h	Largest single N-body run; detailed halo and subhalo catalogs

---

Galaxy Surveys

Completed / Operating

• SDSS (Sloan Digital Sky Survey): began 2000; 2.5m telescope at Apache Point; photometric survey of ~35% of sky + spectroscopy of >4 million objects; discovered BAO signal (2005), mapped cosmic web, largest spectroscopic galaxy catalog for decades
• 2dFGRS (2-degree Field Galaxy Redshift Survey): ~220,000 galaxy redshifts; first BAO-scale detection of galaxy clustering power spectrum features; operated 1997-2002
• DES (Dark Energy Survey): 2013-2019; 5000 sq deg imaging; weak lensing + photometric redshifts; S_8 tension measurements; ~300 million galaxies

Current / Upcoming

• DESI (Dark Energy Spectroscopic Instrument): 2021-2026; 5000-fiber spectrograph on 4m Mayall telescope; targeting 40 million galaxy/quasar spectra; first BAO results released 2024; most precise BAO measurements to date
• Euclid (ESA): launched July 2023; 1.2m space telescope; wide survey ~15,000 sq deg; weak lensing (1.5 billion galaxy shapes) + near-IR spectroscopy (35 million galaxy redshifts); targets dark energy, modified gravity, neutrino masses
• 4MOST (4-meter Multi-Object Spectroscopic Telescope): VISTA telescope, Chile; 2400 fibers; 2024+; cosmology, galactic archaeology, AGN surveys
• Vera C. Rubin Observatory (LSST): first light ~2025; 8.4m; 10-year survey of southern sky; ~20 billion galaxies; weak lensing, photo-z, cluster counts, transients
• Roman Space Telescope (NASA): launch ~2027; 0.28 sq deg near-IR camera; high-redshift BAO + weak lensing; grism spectroscopy for emission-line galaxies; supernova cosmology
• SPHEREx (NASA): launch 2025; all-sky near-IR spectrophotometry; galaxy redshifts for BAO at z < 0.5; ice absorption in star-forming regions

---

Redshift Space Distortions

Galaxy positions in redshift space are distorted by peculiar velocities (non-Hubble motions):

Kaiser Effect (Large Scales)

• Galaxies infalling toward overdensities appear compressed along the line of sight in redshift space
• Enhances apparent clustering: P_s(k) = P_r(k) * (1 + beta * mu^2)^2 where beta = f/b, f is growth rate, b is galaxy bias, mu = cos(theta) angle to line of sight
• Used to measure f * sigma_8 — directly constrains growth of structure and tests gravity theories
• f(z) ~ Omega_m(z)^0.55 in GR (Linder approximation); differs in modified gravity

Fingers of God (Small Scales)

• Random virial motions within galaxy clusters (~500-1500 km/s) elongate structures along line of sight
• Clusters appear as radial "fingers" pointing at observer in redshift cone diagrams
• Must be modeled and corrected in clustering analyses
• Provides velocity dispersion information for cluster mass estimates

---

Power Spectrum and Correlation Functions

Matter Power Spectrum P(k)

• Fourier transform of the two-point correlation function
• Shape encodes: primordial spectrum (nearly scale-invariant, n_s ~ 0.965), transfer function (matter-radiation equality turnover at k_eq ~ 0.01 h/Mpc), BAO wiggles, growth factor
• Turnover scale corresponds to horizon size at matter-radiation equality
• On large scales (k < k_eq): P(k) ~ k^n_s (primordial)
• On small scales (k > k_eq): P(k) ~ k^(n_s - 4) (suppressed by sub-horizon radiation-era growth)
• Normalization: sigma_8 = rms density fluctuation in 8 Mpc/h spheres = 0.811 +/- 0.006 (Planck)

Two-Point Correlation Function xi(r)

• Probability excess of finding a galaxy pair at separation r compared to random: xi(r) = <delta(x) * delta(x+r)>
• BAO bump visible at r ~ 105 Mpc/h (comoving) ~ 147 Mpc
• Galaxy bias: xi_gal(r) = b^2 * xi_matter(r) on large scales

---

CMB Lensing

• CMB photons are gravitationally lensed by intervening large-scale structure (z ~ 0.5-5)
• Smooths acoustic peaks and generates B-mode polarization from E-modes
• Lensing power spectrum C_l^phiphi maps the integrated matter distribution along the line of sight
• Cross-correlation of CMB lensing with galaxy surveys constrains galaxy bias and growth of structure
• Planck lensing amplitude A_L = 1.180 +/- 0.065 — mildly prefers A_L > 1 (the "lensing anomaly", ~2-3 sigma)

---

The Cosmological Principle

Statement

On sufficiently large scales (>200-300 Mpc), the universe is statistically homogeneous and isotropic.

Evidence

• CMB isotropy: temperature uniform to 1 part in 10^5 (after dipole removal)
• Galaxy number counts: approach uniformity on scales >100-200 Mpc
• Fractal dimension approaches 3 (uniform) above ~100 Mpc

Challenges

• Structures like the Sloan Great Wall (~420 Mpc) and Hercules-Corona Borealis Great Wall (~3 Gpc) raise questions about the homogeneity scale
• CMB anomalies: hemispherical asymmetry, cold spot, quadrupole-octupole alignment — statistical significance debated
• Bulk flows: Local Group moves at ~627 km/s relative to CMB; direction and amplitude of flows on larger scales debated
• The principle is statistical, not exact: local fluctuations are expected

---

Anti-Patterns

• Stating galaxy clusters are the largest structures in the universe: Clusters are the largest gravitationally bound structures; superclusters, filaments, and walls are larger but not bound
• Treating superclusters as gravitationally bound systems: Most superclusters will disperse due to cosmic expansion; they are transient features of the current cosmic web
• Confusing comoving and proper distances: BAO scale is ~147 Mpc comoving; proper distances are smaller by factor (1+z) at earlier epochs; always specify which distance measure
• Presenting N-body simulation results as direct observations: Simulations depend on input physics, resolution, and sub-grid models; they are theoretical predictions, not data
• Claiming voids are empty: Voids contain sparse galaxy populations and diffuse gas; they are underdense (~10-20% of mean), not vacant
• Treating galaxy bias as a constant: Bias b depends on galaxy type, luminosity, color, redshift, and scale; it is not universal
• Ignoring systematic errors in BAO measurements: Non-linear evolution, redshift space distortions, galaxy bias, and reconstruction techniques all affect BAO precision
• Conflating the observable universe with the entire universe: We can only observe to z = infinity (comoving distance ~46.5 Gly); the full universe may be much larger or infinite
• Stating that the Great Attractor is a single massive object: It is a region of enhanced density associated with multiple clusters (Norma, Centaurus) and part of the broader Laniakea flow
• Presenting the Fingers of God as a physical structure: They are purely an artifact of peculiar velocities in redshift space; the underlying structure is a compact cluster