Flat Rotation Curves: Weighing a Galaxy's Invisible Halo

The mismatch that would not go away

By the mid-1970s, astronomers measuring how fast stars and gas orbit spiral galaxies kept running into the same puzzle: orbital speeds stayed roughly constant far beyond where the visible starlight thins out, instead of falling off the way Newtonian gravity says they should once you are past most of the mass. Turn off an invisible halo in a rotation-curve model and the predicted speed traces a Keplerian hump that peaks near the edge of the disk and then declines as 1/√r — the same law that governs how fast Mercury orbits compared to Neptune. Real galaxies do not decline; the curve stays flat, sometimes for tens of kiloparsecs past the last bright star.

Isolating the halo's job with a simple model

Strip the problem down to its essentials: a point or extended visible mass Mvis at the centre, plus an extra halo mass Mhalo(r) that grows with radius. Orbital speed on a circular orbit only depends on the mass enclosed inside that radius, v(r) = √(G·M(r)/r), so any component whose enclosed mass keeps growing roughly in step with r will contribute a nearly constant term to v(r) — which is exactly the flat behaviour observed. A halo whose density falls as 1/r² gives enclosed mass M(r) ∝ r exactly, and v(r) becomes perfectly flat; real halos (like the NFW profile) fall off a bit faster at large radius, giving a curve that is flat over a wide range and only gently declines far out, matching real 21 cm radio observations of neutral hydrogen well past the edge of the visible disk.

M(r) ∝ r    (density ∝ 1/r²)   →   v(r) = √(G·M(r)/r) = constant   ← flat curve
M(r) fixed  (all mass inside r) →   v(r) ∝ 1/√r                        ← Keplerian falloff

How much mass is actually missing

Reproduce a flat curve out to, say, 30 kiloparsecs for a typical spiral, and the halo mass needed to make the algebra work out is routinely five to ten times the mass of all the stars and gas combined. That ratio, not a specific particle detection, is the primary quantitative claim behind galactic dark matter: it is inferred purely from applying v(r) = √(GM(r)/r) to a measured curve, the same equation Newton would have used, just solved for the unseen M(r) that the visible matter alone cannot supply.

Why gas at large radius is the crucial probe

Stars alone only trace the rotation curve out to the edge of the visible disk. Cold neutral hydrogen gas, however, is detectable by its 21 cm radio emission far beyond where starlight fades, often two to three times the optical radius. That gas orbits at the same flat speed, which is the real reason the dark matter case is so strong: the flatness is not an artifact of measurement noise fading into darkness at the edge of the disk, it is a clean signal that persists in a completely different tracer, at radii where there is almost no visible mass left to explain it at all under ordinary Newtonian bookkeeping.

What competing explanations have to match

Any alternative to a literal halo of unseen mass — most prominently Modified Newtonian Dynamics, which changes the force law itself below a tiny acceleration threshold instead of adding mass — has to reproduce this same flat behaviour from first principles for every galaxy simultaneously, not just fit each one separately. MOND does remarkably well at the level of individual rotation curves; the harder test is reproducing gravitational lensing by galaxy clusters, the cosmic microwave background's fine structure, and the way large-scale cosmic structure grew over billions of years, all of which favour an actual particle halo. The rotation curve is where the anomaly was first pinned down convincingly; it is not, on its own, proof of what the missing mass is.

Frequently asked questions

Try it live

Everything above runs in your browser — open Galactic Rotation Curve and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.