Module 10: From smooth to lumpy · Lesson 10.3 · about 16 min
Dark matter halos
How a lump of dark matter stops expanding, falls back and settles into a halo about 200 times denser than average, and the NFW shape that every halo shares.
Module 10 · about 16 minutes. Read first: 10.2 Gravity makes lumps. Rotation curves showed that every spiral galaxy sits in a dark halo (Lesson 8.2). Here we see how such halos form, how dense they are, and what shape they take. By the end you will be able to follow a lump from expansion to collapse, give a halo its size from its mass, and read the NFW profile.
1. Up, stop, down: spherical collapse
Picture a ball of matter a little denser than its surroundings. Like a ball thrown upwards (Lesson 5.2), it expands with the universe but slows more than the average, because its extra mass pulls harder. If it is dense enough it stops growing, turns around, and falls back. In 1972 James Gunn and Richard Gott worked this out exactly for a perfectly round ball in a universe of matter alone.
Why half? Once settled, a swarm held by its own gravity obeys the virial theorem: its energy of motion is half the size of its gravitational energy. At turnaround the ball is momentarily still, so all its energy is gravitational; energy is conserved, and the only way to have the same total energy with motion included is to shrink to half the radius. Half the radius means eight times the density of the ball at turnaround; and while it collapsed the universe kept expanding and thinning, which takes the contrast from 5.55 to 18π² ≈ 178.
- Δvir
- the settled halo's density divided by the average density of the universe at that time, for a universe of matter only
- δlin = 1.686
- what the simple growing picture of Lesson 10.2 (δ ∝ a) says the lump's density contrast would be at the moment of collapse, if it had kept growing gently. The real contrast is 178; the 1.686 is a handy threshold: a ripple collapses when its linear δ reaches it
- Physical meaning
- Halos are dense, settled clouds a couple of hundred times denser than the cosmic average. With the threshold 1.686 one can count how many halos of each mass exist at any time, as William Press and Paul Schechter did in 1974.
2. Where does a halo end?
A halo fades smoothly into its surroundings, so astronomers draw an edge by agreement. The commonest choice is r200: the radius inside which the average density is 200 times the critical density (Lesson 5.3), with M200 the mass inside it. Other papers use 200 times the mean matter density (a bigger radius), or the “virial” contrast, which with dark energy today is about 100 times the critical density (Bryan and Norman 1998). The numbers differ by tens of percent, so always check which is meant.
- M200, r200
- the halo's mass and its edge
- ρc
- the critical density today, about 1.3 × 10¹¹ suns per cubic megaparsec (for a halo seen at redshift z, the critical density then)
- Physical meaning
- Mass sets size: r200 grows as the cube root of the mass. The Milky Way's halo, about 10¹² suns, reaches some 200 kpc, more than ten times the radius of its star disc. A rich cluster of 10¹⁵ suns reaches about 2 Mpc.
3. One shape for every halo
When computers could follow millions of particles, Julio Navarro, Carlos Frenk and Simon White looked at the halos that formed, from dwarf galaxies to rich clusters. In 1996 and 1997 they found that all of them had the same shape, now called the NFW profile:
- rs
- the scale radius, where the profile bends
- ρs
- a density that sets the overall level (fixed by the mass and c)
- c
- the concentration: how compact the middle is compared with the edge. Small halos have higher c (about 10 to 20) because they formed earlier, when the universe was denser; clusters have c of about 4 to 6
- Physical meaning
- Inside rs the density rises as 1/r towards the middle (a "cusp"); far outside it falls as 1/r³. In between it falls about as 1/r², which is exactly what makes a rotation curve flat.
Halos are not perfectly round, either: they are slightly squashed, like a rugby ball, and they are full of smaller subhalos, the remains of halos that fell in earlier, some hosting satellite galaxies such as the Magellanic Clouds. The NFW cusp also leads to one of the open puzzles: many small galaxies seem to have smooth “cores” instead (Lesson 8.8).
- 1972 Gunn and Gott: spherical infall and collapse in an expanding universe.
- 1974 Press and Schechter: how many halos of each mass, from the 1.686 threshold.
- 1985 Davis, Efstathiou, Frenk and White's computer models of cold dark matter make halos and galaxy clustering like the real ones.
- 1996-97 Navarro, Frenk and White find one profile for halos of every mass.
- 2008 The Aquarius and Via Lactea simulations resolve Milky Way halos with about a billion particles each, and find tens of thousands of subhalos.
“A dark matter halo is a ring around the galaxy, like an angel’s halo.” It is a huge, roughly round cloud, densest in the middle, with the visible galaxy sitting deep in its centre. Most of the galaxy’s mass is in the halo, spread far beyond the stars; the stars are like the yolk in an egg many times bigger than the yolk.
- A slightly dense ball expands, turns around (5.55 times denser than average), falls back and settles at half its turnaround radius, about 178 times denser than average.
- A ripple collapses when its linear density contrast reaches about 1.686: the threshold for counting halos.
- Halo edges are a convention; r200 encloses 200 times the critical density. M200 = (4π/3) r200³ × 200ρc.
- Halos of every mass share the NFW shape: density ∝ 1/r inside rs, 1/r³ far outside; c = r200/rs.
- The Milky Way’s halo: about 10¹² suns, reaching some 200 kpc, full of subhalos.
Check yourself
Sources
- J. E. Gunn and J. R. Gott (1972), On the infall of matter into clusters of galaxies and some effects on their evolution, ApJ 176, 1 https://doi.org/10.1086/151605
- W. H. Press and P. Schechter (1974), Formation of galaxies and clusters of galaxies by self-similar gravitational condensation, ApJ 187, 425 https://doi.org/10.1086/152650
- J. F. Navarro, C. S. Frenk and S. D. M. White (1996), The structure of cold dark matter halos, ApJ 462, 563 https://doi.org/10.1086/177173
- J. F. Navarro, C. S. Frenk and S. D. M. White (1997), A universal density profile from hierarchical clustering, ApJ 490, 493 https://doi.org/10.1086/304888
- G. L. Bryan and M. L. Norman (1998), Statistical properties of X-ray clusters: analytic and numerical comparisons, ApJ 495, 80 https://doi.org/10.1086/305262
Your ticks stay in this browser. Save them as a file on the course home.
Updated 2026-11-22. Numbers are rounded for teaching; measured values come from the sources listed in each lesson. Simulations run in your browser; nothing you type is sent anywhere.