← all systems · see it in The System
Dark matter halo
The basic unit of cosmological structure: a gravitationally bound region of (mostly dark) matter that has decoupled from the Hubble expansion and collapsed, with a cuspy, nearly universal density profile, a triaxial shape and a small spin, filled with surviving subhalos. Every galaxy forms within one; it envelops the visible galaxy and extends well beyond it. It is not observed directly: it is inferred from rotation curves, the motions of stars, satellites and streams, and gravitational lensing (a modified-gravity reading holds instead that no halo is needed).
Properties published scienceour proposal
Sourced fields, and our own readings.
| born at | first halos: Earth-mass microhalos, first non-linear at z = 60 in a neutralino-CDM simulation (Diemand et al. 2005); minihalos of about 10^6 solar masses at z about 20-30 host the first stars |
| born precision | z about 20-30 for minihalos (model-based) |
| boundary | Inside: the matter bound to the halo and decoupled from the expansion. Where the edge lies is a choice: a radius enclosing a set overdensity (Delta about 333 times the mean, or 200 times the mean or critical density), or the splashback radius where accreted matter reaches its first apocentre (0.8-1.5 R200m), so halos and their effects can extend well beyond the 'virial' radius. |
| nested level | Built from dark matter (and the baryons it holds), with subhalos inside; it sits at a node of the cosmic web, and a galaxy's halo can fall into a group or cluster halo and become a subhalo. |
| scale | galactic |
Measures (73) published science
| characteristic clock: host halo assembly (Milky Way-size, Via Lactea) | most host mass accreted over the first 6; physical mass distribution practically constant since z = 1 Gyr |
| characteristic clock: feedback-driven potential changes | sub-dynamical timescales, repeated over 4 > z > 2 (redshift interval) |
| characteristic clock: self-interaction (hypothesised) | at least one scattering per particle over 10 Gyr for sigma/m about 1 cm^2/g |
| virial overdensity Delta (spherical collapse) | about 333 at z = 0, tending to 178 at high redshift; other authors use 200 (mean or critical): a factor of about 3 times the mean matter density |
| virial mass and velocity: galaxy clusters | Mvir about 10^15, Vvir about 1000 solar masses; km/s |
| virial mass and velocity: Milky Way | Mvir about 10^12, Vvir about 100 solar masses; km/s |
| virial mass and velocity: smallest dwarfs | Mvir about 10^9, Vvir about 10 solar masses; km/s |
| small scales (where the challenges are) | M about 10^11 solar masses, Rvir about 150 kpc, Vvir about 50 km/s, linear scale about 1 Mpc |
| collapsed size against linear scale | Rvir about 7 times smaller than the comoving linear scale of the same mass ratio |
| halo mass function slope | dn/dM proportional to M^alpha with alpha about -1.9 below M*, exponential fall-off above |
| subhalo mass function slope | dN/dm proportional to m^alpha_s with alpha_s about -1.8 |
| most massive subhalo | about 0.2 of the host's mass fraction |
| free-streaming cutoff (100 GeV WIMP) | about 10^-6 solar masses |
| concentration against redshift | c_vir proportional to (1 + z)^-1 at fixed mass |
| concentration scatter | 1 sigma Delta(log c_vir) = 0.18 at given mass (Delta Vmax / Vmax = 0.12) dex |
| dark matter density at galaxy radius across halo masses | rises only by a factor of about 6 over the full mass range of galaxy hosts |
| galaxy size against halo | r_gal about 0.015 Rvir fraction |
| inner slope: simulated cusps against observed cores | gamma about 0.8-1.4 (dark-matter-only) against about 0-0.5 (many dwarfs) log slope |
| halo shape (minor/major axis ratio at 0.3 Rvir) | <s> = (0.54 +/- 0.02) (Mvir/M*)^(-0.050 +/- 0.003); Gaussian scatter about 0.1 |
| Milky Way halo potential flattening (Sagittarius model) | (c/a) = 0.72, (b/a) = 0.99 at 20 < r < 60 kpc; minor axis in the disc plane axis ratio |
| spin parameter distribution | log-normal about lambda'0 = 0.035 +/- 0.005, width 0.50 +/- 0.03 |
| angular momentum alignment | well aligned within each halo for about 80% of halos fraction |
| subhalos resolved (Aquarius, largest run) | nearly 300,000 within the virialized region; up to four generations count |
| mass fraction in substructure | well below 3% within 100 kpc; below 0.1% within the Solar Circle (extrapolated to an Earth mass) fraction |
| subhalo survival since z = 1 (Via Lactea) | 97% keep a bound remnant; Vmax = 10 km/s subhalos keep about 40% of their z = 1 mass fraction |
| splashback radius | about 0.8-1 R200m (rapid accretion) to about 1.5 R200m (slow) |
| steepest outer log slope | -4 and steeper near R200m for massive or rapidly accreting halos log slope |
| galaxy formation efficiency at peak | abstract: less than 20% of available baryons into stars near 10^12 solar masses; body: 'just ∼ 20-30% of baryons have turned into stars' (of the universal baryon fraction, 17%), the stellar-to-halo ratio peaking at a few percent |
| stellar mass scatter at fixed halo mass (above pivot) | less than 0.2 dex |
| disc mass against halo | a few percent; specific angular momentum similar to the halo's fraction |
| adiabatic contraction model error | standard model overpredicts in the inner 5% of Rvir; modified model within 10-20% fraction |
| Milky Way satellites against subhalos | about 50 known galaxies (stellar mass > 300 solar masses) within 300 kpc against up to about 1000 dark subhalos above 10^7 solar masses count |
| Local Group satellites 1999 (Klypin) | about 50 predicted above 20 km/s in 570 kpc against about a dozen observed; about 300 predicted in 1.5 Mpc against about 40 count |
| too-big-to-fail subhaloes | peak circular velocities at infall 30-70 km/s; infall masses 0.2-4 x 10^10 solar masses; density would need to fall by order unity at 0.3-1 kpc km/s; solar masses |
| Milky Way virial mass (Kafle 2014) | 0.80 (+0.31/-0.16) x 10^12 solar masses |
| Milky Way concentration (Kafle 2014) | 21.1 (+14.8/-8.3) |
| local dark matter density | 0.0088 (+0.0024/-0.0018) solar masses per cubic parsec (0.35 GeV/cm^3) Msun/pc^3 |
| Milky Way mass within 50 kpc | 3.80 (+0.14/-0.11) x 10^11 solar masses |
| LMC total mass | 1.38 (+0.27/-0.24) x 10^11 solar masses |
| Milky Way virial mass in the LMC-wake model | 1.2 x 10^12 (McMillan 2017), Delta_vir = 357 solar masses |
| Milky Way mass measurements scatter | a factor of two over the last two decades |
| Milky Way dark fraction (encyclopedia) | about 95% of the galaxy; halo 6 x 10^11 to 3 x 10^12 solar masses |
| minihalos (first-star hosts) | total masses about 10^6 solar masses collapsing at z about 20-30 |
| self-interaction cross-section (Spergel and Steinhardt range) | 0.45-450 cm^2/g |
| self-interaction cross-section to solve small-scale issues | about 0.5-1 cm^2/g |
| fuzzy dark matter particle mass | about 10^-22 eV |
| radial acceleration relation scale g-dagger | 1.20 +/- 0.02 (random) +/- 0.24 (systematic) x 10^-10 m/s^2 |
| radial acceleration relation sample | 2693 points in 153 galaxies count |
| MOND residual missing mass in clusters | roughly a factor of two |
| Bullet Cluster lensing-baryon offset | 8 sigma (z = 0.296) significance |
| lopsidedness against potential depth | Spearman -0.432 (A1 vs gas rotation, observed); -0.39 (A1 vs Vmax, TNG50) rank correlation |
| first halos (neutralino CDM) | Earth-mass haloes about as large as the Solar System, first non-linear at z = 60 in the simulation; over 10^15 expected to survive in the Galactic halo solar masses (Earth mass) |
| SIDM core collapse time | t_coll about 290 t_r (relaxation times), longer than a Hubble time for a typical dark-matter-dominated galaxy core today |
| inner slope against stellar-to-halo mass (feedback) | cusp at M*/Mhalo about 0.01%; flattest profiles at about 0.5%; cuspier again above fraction |
| warm dark matter mass limit (satellite census) | m_WDM > 6.5 keV (95%) |
| fuzzy dark matter mass limit (satellite census) | m_phi > 2.9 x 10^-21 eV (95%) |
| warm dark matter in tension below (completeness-corrected counts) | thermal relic mass below 4 keV in tension with satellite counts |
| MACHO halo fraction (EROS-2) | less than 8% of the halo mass at about 0.4 solar masses; 0.6 x 10^-7 to 15 solar masses ruled out as primary occupants fraction |
| LMC wake detected | local wake and collective response in 1301 stars at 60-100 kpc; density contrasts stronger than in the simulation count |
| Milky Way reflex motion (disc against outer halo) | v_travel = 32 (+4/-4) km/s against tracers at 40-120 kpc |
| LMC mass from the Sagittarius stream | (1.3 +/- 0.3) x 10^11 solar masses |
| Milky Way mass from the Sagittarius stream (with the LMC) | within 100 kpc (5.6 +/- 0.4) x 10^11; virial (9.0 +/- 1.3) x 10^11 solar masses |
| Milky Way mass bias if the LMC is ignored | overestimated by as much as 50%; about 10^12 solar masses after accounting for a 1.5 x 10^11 LMC fraction |
| Milky Way total mass (Keplerian-decline reading) | 2.06 (+0.24/-0.13) x 10^11, strict upper limit 5.4 x 10^11; rotation falls about 30 km/s between 19.5 and 26.5 kpc; flat curve rejected at 3 sigma solar masses |
| shock-heating mass | M_shock about 10^12 solar masses |
| splashback detection (SDSS clusters) | strong evidence; radii smaller than N-body expectations; assembly bias detected at 6.6 sigma, stronger than expected |
| Gaia-Enceladus merger | infall slightly more massive than the SMC; 4:1 mass ratio; about 10 Gyr ago |
| GD-1 perturber mass | 10^6 to 10^8 solar masses; inferred density 2-3 sigma above LCDM subhalo expectations there |
| dark satellite by lensing (JVAS B1938+666) | 1.9 +/- 0.1 x 10^8 solar masses at z = 0.881; substructure mass fraction 3.3 (+3.6/-1.8)% |
| bar stability with a halo (1973) | halo-to-disc mass ratios of 1 to 2 required for stability; t about 0.14 |
| reionisation suppression scale | gas accretion suppressed below Mvir about 10^9 solar masses (Vmax about 30 km/s) |
| standard halo model refined (SHM++) | five parameters: circular speed, escape speed, local density, anisotropy and the sausage's dark matter density; modest impact on WIMP signal models |
| inner profile shape (Einasto) | better described by the three-parameter Einasto profile; subhalo inner profiles curve gently with no fixed asymptotic slope |
Open contentions (12) open contention
Where sources or readings disagree. Both stay on the map.
- C-098 Are galaxy rotation and cosmic acceleration caused by unseen matter and energy, or by gravity behaving differently?
booked: The map follows the mainstream view of unseen components (dark matter and dark energy), and keeps modified gravity alongside.
the other reading: modified dynamics: Milgrom's law with a0 of order the square root of the cosmological constant predicts many galaxy observations (Famaey & McGaugh 2012); the radial acceleration relation, 'tantamount to a natural law', fixes the dark contribution from the baryons (McGaugh et al. 2016); both (Famaey & McGaugh's option iii) - C-159 Why do some galaxies' dark-matter centres look flat instead of peaked?
booked: The map keeps the competing explanations side by side for now.
the other reading: baryonic feedback: repeated potential fluctuations (Pontzen & Governato 2012); self-interacting dark matter at about 0.5-1 cm^2/g (Tulin & Yu 2018); ultra-light fuzzy dark matter, m about 10^-22 eV (Hu et al. 2000); warm dark matter (named in the encyclopedia; not read here); Oman et al. 2015: better called an 'inner mass deficit' problem with diverse rotation curves at fixed Vmax; one or more must hold: '(i) the dark matter is more complex than envisaged by any current model; (ii) current simulations fail to reproduce the effects of baryons on the inner regions of dwarf galaxies; and/or (iii) the mass profiles of "inner mass deficit" galaxies inferred from kinematic data are incorrect'; feedback's measured curve (Di Cintio et al. 2014): cusps at M*/Mhalo 0.01%, flattest at 0.5%, cuspier above - C-3373 What makes disc galaxies lopsided?
booked: The map keeps the competing readings; none is ruled.
the other reading: tidal interactions and minor mergers; asymmetric cold-gas accretion; internal collective instabilities (Omonov et al. 2026 read the observed m = 1 amplitude as regulated by internal dynamics) - C-3604 Does a dark matter halo prevent a galaxy's bar or help it grow?
booked: The map keeps the competing readings; none is ruled.
the other reading: a halo stabilises cold discs against bars (Ostriker & Peebles 1973: halo-to-disc mass 1-2 needed); a live halo's resonant stars 'can stimulate, rather than restrain, the growth of the bar' (Athanassoula 2002) - C-3605 Where does a dark matter halo end?
booked: The map keeps the competing readings; none is ruled.
the other reading: the splashback radius as the physical boundary (More, Diemer & Kravtsov 2015), at 0.8-1.5 R200m; overdensity boundaries give 'pseudo-evolution' - C-3606 How light can a dark matter particle be if dark matter is 'fuzzy'?
booked: The map keeps the competing readings; none is ruled.
the other reading: the Milky Way satellite census sets m > 2.9 x 10^-21 eV at 95% (Nadler et al. 2021), in tension with the original value - C-3607 Why does the Milky Way have so few small satellite galaxies: dark matter physics, or gas that never made stars?
booked: The map keeps the competing readings; none is ruled.
the other reading: dark matter physics suppresses small halos (warm or fuzzy dark matter); satellites missed by surveys, or identified with high-velocity clouds (Klypin et al. 1999 listed these); number counts solved after completeness corrections, with a possible 'too many satellites' problem (Kim, Peter & Hargis 2018); the DES census is 'consistent with the cold, collisionless DM paradigm' (Nadler et al. 2021) - C-3608 How massive is the Milky Way's dark matter halo?
booked: The map keeps the competing readings; none is ruled.
the other reading: measurements over two decades scatter by a factor of two; Gaia DR2-era values approach higher precision (Wang et al. 2020); Vasiliev et al. 2021: virial (9.0 +/- 1.3) x 10^11; Erkal, Belokurov & Parkin 2020: ignoring the LMC overestimates the mass by as much as 50%; about 10^12 after correction; Jiao et al. 2023: a Keplerian decline beyond 19 kpc; total 2.06 x 10^11, upper limit 5.4 x 10^11 - C-3609 What shape is the Milky Way's dark matter halo?
booked: The map keeps the competing readings; none is ruled.
the other reading: the LMC's pull may remove the need for triaxiality (Law & Majewski's own alternative; Erkal et al. 2019: a static Milky Way cannot be assumed); the halo's orientation evolved with time; Vasiliev, Belokurov & Erkal 2021: LMC (1.3 +/- 0.3) x 10^11 solar masses; Galaxy lurches toward the Cloud - C-3611 Do the thin planes of satellite galaxies contradict the standard picture of dark matter?
booked: The map keeps the competing readings; none is ruled.
the other reading: the Milky Way's satellites on a polar plane cannot be cosmological substructure (Kroupa et al. 2005, as Bullock & Boylan-Kolchin report); subhalo distributions are themselves highly non-spherical and the plane is marginally consistent (Zentner et al. 2005, as reported) - C-3612 Do galaxy clusters end where simulations say they should?
booked: The map keeps the competing readings; none is ruled.
the other reading: the measured radii are smaller and the assembly bias stronger than N-body expectations; dynamical friction or cluster-finding systematics (miscentering, projection) named as possible causes (More et al. 2016) - C-3613 Why are the Milky Way's largest satellites less dense than dark matter simulations predict?
booked: The map keeps the competing readings; none is ruled.
the other reading: baryonic feedback lowers central densities of massive subhaloes; different dark matter physics (e.g. self-interaction); the Milky Way is a statistical anomaly; galaxy formation becomes stochastic at 30-70 km/s
Cross-scale patterns seen here (18) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Dark matter halo. Each links to every other system that shows it.
- same role, different builds (structural)
One job, turning a dense central cusp into a flat core, done by three unrelated builds: gas-driven potential fluctuations (baryonic feedback), particle self-scattering (SIDM, about 0.5-1 cm^2/g) and the wave pressure of ultra-light 'fuzzy' dark matter. The same holds one floor up: the unseen pull th - the whole sets the local reference (structural)
Each halo's edge and mass are measured against the whole universe's density (Delta times the mean or critical density), so a halo appears to grow just because the reference density falls ('pseudo-evolution') while its physical mass distribution stays constant since z = 1; the choice of reference shi - Umwelt: each system carves its own slice of one shared medium (structural)
The halo's channels are all gravity (and, if dark matter self-interacts, collisions among its own particles): it 'interacts extremely weakly with atoms and photons, so that gravity alone has affected its distribution'. The baryons sharing the same volume hear light, pressure and heat; the halo does - conservation law plus closure (partial)
Virial equilibrium leaves the halo's mass open: 'the value of Mvir is ultimately a definition that requires some way of defining a halo's outer edge', closed by choosing Delta (or the splashback radius). Flat rotation curves likewise need a second law to close them: a dark mass profile with Newton, - holding / entrainment (partial)
The host sets the count and fate of its parts: subhalo counts are nearly self-similar with host mass, tides strip subhalos and dynamical friction drags them in; but 97% keep a bound remnant, so the hold is on mass, not life or death. - lightning (partial)
Repeated energetic starbursts discharge into the centre: the potential 'changes on sub-dynamical timescales', and what follows is not the same (a core that dwarfs keep to the present). The discharge is the gas's; the halo is what is changed. - not-diluting wins (partial)
A halo decoupled from the expansion keeps its physical mass distribution 'practically constant since z = 1' while the background density falls, so its contrast keeps rising; cold, pressureless dark matter clumps first because it 'removes most of the thermal and radiative pressures' holding the baryo - runaway from an unstable balance (partial)
Gravitational instability: perturbations grow until they reach a critical density, then 'stop expanding, turn around, and collapse'; past turnaround the region leaves the expansion for good. No negative heat capacity is stated in the sources read. - same shape, opposite sign (partial)
One process, the dark matter's orbits answering a change in the central baryonic mass, run both ways: gas condensing into the centre contracts the dark matter (denser, steeper), gas blown out of the centre in repeated bursts cores it; and 'if too many baryons end up in stars, the excess central mass - smooth to sharp in finite time (partial)
If dark matter self-interacts, the core 'undergoes secular collapse to a singular state (the "gravothermal catastrophe") in a time t_coll ≈ 290 t_r': a smooth evolution reaching a singular value at a finite time (hypothesised system). - spin-up on contraction (ice skater) (partial)
The halo holds a small spin (lambda' about 0.035) from tidal torques; gas cooling inside it keeps a specific angular momentum similar to the halo's and so ends rotationally supported in a disc a few percent of the halo's mass, the spin halting the collapse. - symmetry breaking (name the cut) (partial)
Collapse from a nearly uniform field is aspherical: accretion is 'directional and tends to be clumpy', so halos come out triaxial (mean minor/major about 0.54), with axes set by where matter arrived from; in the Milky Way a fit puts the minor axis in the disc plane (contested). - templating (partial)
One profile reused across scales: the spherically averaged profiles 'of all our halos can be fit over two decades in radius by scaling a simple universal profile', from dwarf halos to rich clusters, and substructure counts are nearly self-similar with host mass. The halo itself is the reused module: - The carry (partial)
The wait: 'The growth rate of halos controls the rate at which baryons are accreted', so a galaxy's supply waits on its halo's growth; inside the halo, accreted matter overshoots to its first apocentre (splashback) before settling, and late halos keep 'quiet' interiors while the outskirts still fill - The Payment to Entropy (partial)
Dynamical-thermal form: when a satellite moves through the halo its orbital energy is handed to halo particles as a wake and its orbit decays; when feedback shakes the centre, fluctuations 'irreversibly transfer energy into collisionless particles'. The ordered energy (an orbit, a gas flow) arrives - who opens the door (partial)
Most crossings into the halo are open field: a satellite's wake, lensing and contraction are gravity acting on whatever particles are there. The one prepared receiver read is self-built by the halo's own physics: halo orbits at resonance with a bar, whose capacity to absorb 'depends on the distribut - borrowed-then-held (surface)
Spin borrowed from neighbours' tidal torques, then held in the halo (well aligned in about 80%) and handed to the disc; a conserved quantity, not a regulation, hence surface. - partner-becomes-part (surface)
A separate halo falling into a larger one 'becomes a subhalo orbiting within the potential well of its host'; there is no gain/gain loop before the fusion (the host strips it), so surface.
Valence: what it trades, and with whom (17) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Barred spiral galaxy via sys:gravitational-field · Dark matter halo gain
- Photon via sys:gravitational-field · Dark matter halo even
- Supernova via sys:gravitational-field · Dark matter halo mixed
- Dark matter via sys:gravitational-field · Dark matter halo gain
- Interstellar medium via sys:gravitational-field · Dark matter halo mixed
- Galaxy via sys:gravitational-field · Dark matter halo mixed
- Galaxy cluster via sys:gravitational-field · Dark matter halo gain
- Human (Homo sapiens) via sys:gravitational-field · Dark matter halo even
- Subhalo via sys:gravitational-field · Dark matter halo gain
- Cosmic web via sys:gravitational-field · Dark matter halo gain
- Circumgalactic medium (CGM) via sys:gravitational-field · Dark matter halo even
- Population III star via sys:gravitational-field · Dark matter halo even
- Galactic disc via sys:gravitational-field · Dark matter halo mixed
- Sagittarius dwarf galaxy via sys:gravitational-field · Dark matter halo even
- Large Magellanic Cloud via sys:gravitational-field · Dark matter halo mixed
- Lopsidedness of galactic discs (the m = 1 mode) via sys:gravitational-field · Dark matter halo even
- Galactic bar via sys:gravitational-field · Dark matter halo gain
Partners and parts published science
Built from / builds (8)
- Galaxy
- Galaxy cluster
- Dark matter
- Baryonic matter
- Subhalo
- Circumgalactic medium (CGM)
- Host halo against its subhalos for the subhalos' mass
- Angular momentum sinks against the growing centre, for the bar's strength
Sits inside / holds (8)
- Primordial black hole
- Barred spiral galaxy
- Self-interacting dark matter
- Galaxy cluster
- Cosmic web
- Lambda-CDM model
- Galactic disc
- Lopsidedness of galactic discs (the m = 1 mode)
expressed by (4)
- Recombination/Composition
- Scaffolding/Enablement (Meta-Tooling)
- Niche Construction (alter environment)
- Resource Denial/Attrition
floor filled by (23)
- Dark matter
- Molecular hydrogen (H2, dihydrogen)
- Dark matter-baryon partnership (structure formation)
- Competition over the expansion (dark energy vs matter's gravity)
- Galaxy cluster
- Barred spiral galaxy
- Self-interacting dark matter
- Photon
- Supernova
- Interstellar medium
- Galaxy
- Human (Homo sapiens)
- Subhalo
- Cosmic web
- Circumgalactic medium (CGM)
- Population III star
- Lambda-CDM model
- Galactic disc
- Large Magellanic Cloud
- Stellar halo
- Galactic bar
- Caustic of a timelike congruence (crossing of free-falling matter world lines)
- Lopsidedness of galactic discs (the m = 1 mode)
represents (1)
Kinds (12)
- Subhalo
- Cold dark matter
- Warm dark matter
- Self-interacting dark matter
- Massive compact halo object (MACHO)
- MOND (relativistic forms)
- Fuzzy cold dark matter
- Minihalo
- Earth-mass microhalo
- Cluster halo
- Dwarf galaxy halo
- Dark (starless) halo
wired (9)
- Galaxy
- Dark matter constituent (unidentified; hypothesised WIMP or axion)
- Subhalo
- Self-interacting dark matter
- Star
- Interstellar medium
- Galaxy filament
- Intracluster medium (ICM)
- Galactic bar
Notes published science
- A galaxy can be pulled lopsided by something you can't see: a few percent asymmetry in the dark matter halo is enough to make the disc about 10% lopsided, so astronomers can read the invisible halo's shape from the visible disc, as long as the disc has settled. Our own Milky Way has not: its disc is moving at about 32 km/s relative to the outer halo, knocked off-centre by the Large Magellanic Cloud (src:jog-combes-2009, src:petersen-penarrubia-2021).
- The Large Magellanic Cloud weighs about 1.4 x 10^11 suns, enough to bend the Orphan stream and swing the inner Milky Way towards it. The wake it drags through our dark halo was predicted in 2019 and seen in 2021 in 1301 distant halo stars, stronger than the simulation, and the Milky Way's disc was caught moving at about 32 km/s against the outer halo: the halo is not sitting still (src:erkal-2019-lmc, src:conroy-2021-lmc-wake, src:petersen-penarrubia-2021).
- Supernovae cannot touch dark matter, yet they can move it: repeated bursts blow the central gas out and let it fall back, the gravity at the centre shakes faster than the dark matter can follow, and the energy pumped in turns a dense cusp into a flat core, even when only a few per cent of the gas becomes stars (src:pontzen-governato-2012).
- Where a halo ends is partly a bookkeeping choice: measured against the universe's average density, a halo seems to keep growing simply because the universe thins around it ('pseudo-evolution'), while its own mass has hardly moved since z = 1. Its physical edge, where infalling matter turns round for the first time, can lie about 1.5 times further out (src:wechsler-tinker-2018, src:diemand-kuhlen-madau-2007, src:more-diemer-kravtsov-2015).
- A bar gets stronger by giving its spin away. Inside corotation the bar, as a pattern, carries negative energy and angular momentum, so every bit of angular momentum it hands to the dark matter halo and outer disc makes it grow; a halo that soaks up more makes a stronger, slower bar (src:athanassoula-2012-bars-review).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
baryons in the well (gas cooling, star formation and feedback at the centre)
- L1: Gas in the well: the halo's gravity overcomes the gas's thermal energy so it collapses into stars and galaxies
- L2: Cooling and condensation: dissipating gas settles into a rotationally supported disc a few percent of the halo's mass, with similar specific angular momentum
- L3: Bursts and outflows: centrally concentrated bursts of star formation blow large underdense bubbles; galactic fountains
- L4: The dark centre answers: contracted by condensing gas (less than the standard adiabatic model predicts) or cored by repeated potential fluctuations; too many stars drag dark matter back in
electromagnetic field (light from behind and within: lensing and tracers observers read)
- L1: Light from background galaxies, and from stars and gas orbiting in the halo
- L2: Paths bent by the halo's mass: weak and strong lensing (giant arcs in clusters); 21-cm and stellar lines Doppler-shifted by orbits in the well
- L3: Read as a mass map: the lensing potential traces the galaxies, not the X-ray plasma, in a cluster merger
- L4: Set against the baryons: the radial acceleration relation ties the observed acceleration to that of the visible baryons even where dark matter dominates
gravitational field (the halo's potential: what orbits in it and what it exchanges with)
- L1: Mass that only gravitates: a particle that interacts extremely weakly with atoms and photons, so that gravity alone has shaped its distribution
- L2: The potential well: circular velocity that stays flat with radius beyond the visible disc
- L3: Tracers on orbits: halo stars, globular clusters, satellites and tidal streams whose motions weigh the halo and map its shape
- L4: Exchanges with what orbits: density wakes behind satellites (dynamical friction); angular momentum absorbed at resonance from bars
- L5: The halo's response: Transient and Collective wakes, the reflex motion of the inner Galaxy towards the LMC, a halo no longer static; observed: the wake and collective response in outer-halo stars, the reflex motion, an outer halo out of equilibrium
the boardroom (shared floors)
- L4: The virialized body: bound, decoupled from expansion; mass, radius and velocity at a chosen overdensity, or at the splashback edge; reached by violent (collisionless) relaxation in a rapidly changing potential
- L5: Internal structure: cuspy near-universal profile, concentration, triaxial shape, spin; contracted by gas, cored by feedback or (hypothesised) self-interaction; if dark matter self-interacts, a core that can later collapse (gravothermal catastrophe, hypothesised)
- L6: Substructure and assembly history: subhalos stripped and sinking; merger history setting concentration and clustering
- L7: Host and pull: the galaxy it holds (efficiency by halo mass), rotation curves, lensing, satellites' decaying orbits
the dark matter halo as cosmologists and galactic dynamicists describe it (reader's view)
- L1: The setting: a cosmic density field seeded by primordial adiabatic fluctuations in an expanding background; halos form within nodes of the cosmic web
- L2: Growth: perturbations grow by gravitational instability, the smallest scales going non-linear first; the halo mass function follows from the linear power spectrum (Press-Schechter and its extensions)
- L3: Turnaround and collapse: an overdense region stops expanding, turns around and collapses, potential energy turned into kinetic; infalling matter overshoots to its first apocentre (the splashback radius)
- L4: The virialized body: a gravitationally bound clump decoupled from the Hubble expansion, labelled by mass, radius and circular velocity at a chosen overdensity (Delta about 333 at z = 0) or by its splashback edge
- L5: Internal structure: a cuspy, nearly universal density profile (NFW or Einasto) whose concentration falls with mass and reflects formation time; triaxial shapes; a spin parameter near 0.035; cusps contracted by cooling gas or flattened into cores by feedback (or by self-interaction, hypothesised)
- L6: Substructure and assembly: smaller halos survive inside as subhalos (up to four generations), stripped from the outside in and sinking by dynamical friction; the merger and accretion history sets concentration and large-scale clustering (assembly bias)
- L7: Host and pull on the world: every galaxy forms within a halo, with a star-formation efficiency set by halo mass (peaking near 10^12 solar masses, under 20% of baryons into stars); the halo's pull keeps rotation curves flat, bends background light and decays satellites' orbits
way in: Dark matter scattering off itself (hypothesised: self-interacting dark matter)
- L1: Dark matter particles in a dense centre (about 0.1 solar masses per cubic parsec, relative speeds about 50 km/s in a dwarf)
- L2: Scattering, if the cross-section per unit mass reaches about 1 cm^2/g: at least one scatter per particle in 10 Gyr
- L3: Heat conduction in the inner halo
- L4: Cores that would solve core-cusp and too-big-to-fail at 0.5-1 cm^2/g while leaving large scales intact; in the long-mean-free-path regime the core is a phase: it later collapses gravothermally to a singular state in about 290 relaxation times
way in: Feedback: bursts of star formation and supernova-driven outflows
- L1: Centrally concentrated bursts of star formation and their supernovae
- L2: Large underdense bubbles of expanding gas; outflows and fountains
- L3: The central kiloparsec's potential changes on sub-dynamical timescales
- L4: Energy irreversibly transferred into the collisionless dark matter: a core
way in: Gas cooling and condensing at the centre
- L1: Gas pulled into the well, its thermal energy overcome
- L2: Dissipation: the gas cools and sinks to the centre
- L3: A central disc a few percent of the halo's mass, the halo responding adiabatically (the model's assumption)
- L4: Dark matter pulled inward: denser, steeper inner profile, but less than the standard adiabatic-contraction model in the inner 5% of the virial radius
way in: Infall of diffuse matter and of smaller halos (mergers)
- L1: Diffuse matter and smaller halos along filaments of the cosmic web
- L2: Infall that is directional and clumpy; most of a Milky Way-size halo's mass arrives in the first 6 Gyr in major mergers
- L3: First apocentre after turnaround: the splashback radius, a sharp density drop placed by the accretion rate
way in: Satellites and infalling galaxies (the LMC, Sagittarius)
- L1: An orbiting satellite's mass (the LMC about 1.4 x 10^11 solar masses)
- L2: Direct gravitational scattering of halo particles
- L3: A density wake: the Transient response trailing the orbit, the Collective response across the halo (both detected in 1301 outer-halo stars, Conroy et al. 2021)
- L4: Reflex: the inner Milky Way moves about the barycentre with the LMC (detected: the disc moves at about 32 km/s against outer-halo tracers)
- L5: The satellite loses energy and angular momentum to the halo particles and is stripped from the outside in
way in: The disc and its bar's non-axisymmetric pull
- L1: The disc inside the halo, a few percent of its mass
- L2: A bar emits angular momentum from inner-disc material at resonance
- L3: Resonant halo material absorbs it, by an amount set by its distribution function and velocity dispersion
- L4: The halo (with the bulge) holds the angular momentum the bar lost; the bar grows stronger and slower
way in: The surrounding density field and the expansion
- L1: Primordial adiabatic fluctuations in an expanding background
- L2: Linear growth by gravitational instability; the smallest scales go non-linear first
- L3: Turnaround: the shell stops expanding and collapses
way in: Tidal torques from neighbouring structure
- L1: The tidal field of neighbouring structure on a collapsing region
- L2: Asymmetric collapse in an expanding universe picks up angular momentum (tidal-torque theory), and minor mergers carry more in
- L3: Spin acquired: spin parameters log-normal about 0.035
- L4: Spin held by the virialized body: angular momentum cylindrical and well aligned in about 80% of halos
way out: An angular-momentum sink for bars
- L1: The disc's stars on bar orbits
- L3: The bar grows stronger and slower
- L5: Resonant halo material absorbs the bar's angular momentum
way out: Annihilation light (hypothesised)
- L1: Gamma rays (none confirmed in the sources read)
- L2: Particle-particle annihilation to gamma rays, if dark matter annihilates
- L6: Dense subhaloes and Earth-mass microhalos as the brightest expected sources
way out: Dynamical friction and tides on satellites
- L1: The satellite stripped from the outside in, most mass lost at first pericentre
- L3: Orbits decay over time
- L6: The wake pulls back on the satellite
way out: The pull on what orbits in it, and on light
- L1: Stars, gas, satellites and light paths in the well
- L4: Light from behind bent into arcs and shear
- L7: The halo's mass keeps rotation curves flat beyond the visible disc
way out: The virial shock: cold streams below 10^12 solar masses, hot dilute gas and a cut-off supply above
- L1: Below M_shock cold streams reach and build the disc
- L4: A stable virial shock heats infalling gas in haloes above about 10^12 solar masses
- L7: Above M_shock the hot, dilute gas becomes vulnerable to AGN feedback; the supply shuts off and galaxies turn red and dead
way out: The well that lets gas cool into galaxies
- L1: The first stars in minihalos of about 10^6 solar masses at z about 20-30
- L3: Gravity overcomes the gas's thermal energy and it collapses
- L7: Galaxy formation efficiency set by halo mass: under 20% of baryons into stars at the 10^12 solar-mass peak