Galaxy Strangulation: Cutting Off the Cold-Gas Supply

What strangulation is and why it matters

Star formation is a supply problem. A galaxy converts cold molecular gas into stars over a depletion time of only ~1–2 Gyr, so to keep forming stars for a Hubble time it must be continuously refueled — by cooling from a hot gaseous halo and by cold cosmic accretion streaming in along filaments. Strangulation is the loss of that resupply: cut the inflow, and the galaxy is left to consume whatever cold gas it already holds and no more.

The name captures the slow, suffocating character of the process. Unlike a sudden strip, nothing dramatic happens to the disk at the moment of cutoff. The galaxy keeps shining and forming stars, seemingly normal, while its reserve quietly drains. This makes strangulation a leading explanation for environmental quenching — the observation that galaxies falling into groups and clusters are far more likely to be red and dead than isolated 'field' galaxies of the same mass. It is one of the central mechanisms shaping the sharp division of galaxies into a blue star-forming cloud and a red passive sequence.

The mechanism, step by step

Consider a star-forming galaxy falling into a larger halo as a satellite. In isolation it sits at the center of its own dark-matter halo with a hot gas atmosphere (~10⁶–10⁷ K) that radiatively cools onto the disk, plus cold accretion from the cosmic web.

On infall, two things sever the resupply. First, the satellite is tidally truncated: it no longer sits at a halo center, so cosmological cold accretion that once fed it is redirected to the central galaxy. Second, its own hot atmosphere is removed — stripped by ram pressure against the group/cluster intracluster medium and by tidal forces — so cooling-driven refueling stops too. The disk's dense cold gas, held deeper in the potential, initially survives.

With inflow choked off, the galaxy keeps forming stars from its remaining disk gas. Star formation itself consumes gas and drives modest outflows, so the reservoir declines on the gas-depletion timescale. Star-formation rate fades gradually until the cold gas is exhausted and the galaxy quenches — a slow strangling rather than a quick kill.

Characteristic numbers, scales, and the metallicity criterion

The defining number is the quenching timescale: Peng, Maiolino & Cochrane (2015) inferred a total time from cutoff to full quiescence of about 4 Gyr for galaxies around 10¹⁰–10¹¹ M☉, longer than the ~1–2 Gyr gas-depletion time because low-level accretion and recycling linger. This is far slower than ram-pressure stripping (~0.1–1 Gyr).

Their key diagnostic is stellar metallicity. In a healthy galaxy, infalling low-metallicity gas continuously dilutes the interstellar medium, keeping metallicity down as stars enrich it. Cut off that pristine inflow and the closed-box system keeps enriching without dilution, so metallicity climbs. Peng et al. found that local passive galaxies have systematically higher stellar metallicity than star-forming galaxies of the same mass — the gap growing toward lower mass — exactly the fingerprint of strangulation. Quantitatively, matching the observed metallicity offset (of order +0.1 dex) requires the star-forming population to have been starved for a few Gyr before quenching, pinning the ~4 Gyr number.

How it is observed and detected

Strangulation leaves no single dramatic image, so it is inferred from population statistics and chemistry. The landmark evidence comes from the Sloan Digital Sky Survey (SDSS): from ~26,000 local galaxies, Peng et al. (2015) measured stellar metallicities from optical absorption-line spectra and found the passive-vs-star-forming metallicity offset that a slow-starvation model predicts, while a rapid outflow-only model does not.

Complementary signatures come from gas surveys. H I 21 cm observations (e.g. ALFALFA, VIVA in Virgo) show satellites that are H I-deficient yet still hold molecular gas and disturbed-but-present disks — consistent with resupply loss rather than wholesale stripping. Integral-field surveys such as MaNGA and SAMI map spatially resolved star formation, showing satellites fading gradually and often from the outside in. Studies like K-CLASH probe strangulation and ram pressure together in clusters at z ≈ 0.3–0.6, and phase-space diagrams (velocity vs cluster-centric radius) separate recently-infallen from long-resident satellites to time the quenching.

Where it operates, and how it differs from stripping

Strangulation is primarily a satellite phenomenon: it needs a larger host halo whose environment can strip the satellite's own atmosphere and steal its accretion. It operates across galaxy groups and clusters — even loose groups suffice, since removing the tenuous hot halo is far easier than tearing out dense disk gas. Peng et al. argued it dominates quenching for galaxies below ~10¹¹ M☉ across cosmic time.

The crucial distinction is what is removed and how fast. Ram-pressure stripping directly tears the cold ISM out of the disk when ρ_ICM·v² exceeds the disk's gravitational restoring force per area — fast (~10⁸ yr), producing one-sided gas tails and 'jellyfish' galaxies, dominant near cluster cores at high speed. Strangulation only removes the outer reservoir, is slow (Gyr), and leaves the disk morphologically intact until fuel runs out. In practice the two act in sequence: strangulation begins on infall in the outskirts, and ram pressure delivers the coup de grâce deeper in. Tidal harassment and internal AGN/stellar feedback add further channels, the latter also quenching massive centrals where strangulation does not apply.

Open questions and significance

Strangulation reframed environmental quenching: because it is slow and gentle, it naturally explains why many cluster satellites are red yet structurally still disk-like, and why passive galaxies carry a chemical memory of gas cutoff. It ties a galaxy's fate to when and how it joined a larger halo.

Open questions remain. How is quenching partitioned between slow strangulation and fast ram pressure, and does that mix shift with cluster mass, redshift, and orbit? The classic ~4 Gyr timescale is a population average — resolved data hint at 'slow-then-rapid' quenching, where a long starvation phase ends in a brief stripping-driven plunge. It is also debated whether strangulation reaches lower-mass centrals via 'cosmic web detachment' or preventive feedback, not just satellites. And simulations (EAGLE, IllustrisTNG) must reproduce both the metallicity offsets and gas-deficiency statistics simultaneously — a demanding joint test of how galaxies are refueled and cut off. Resolving these will sharpen our picture of why star formation across the Universe has been winding down for the last ~10 Gyr.

Strangulation compared with the other main environmental quenching channels for satellite galaxies

Process	What it removes	Timescale	Key signature
Strangulation / starvation	Extended halo gas (hot atmosphere + cold accretion); disk gas untouched at first	Slow, ~2–4 Gyr	Elevated stellar metallicity in passive galaxies; gradual color reddening
Ram-pressure stripping	Cold ISM disk gas directly, stripped by intracluster medium	Fast, ~0.1–1 Gyr	One-sided gas tails, 'jellyfish' galaxies (Hα, H I, radio)
Tidal / harassment	Stars + gas via gravitational interactions and high-speed encounters	Varies, ~1 Gyr per pass	Tidal tails, disturbed morphology, bridges
AGN / stellar feedback	Gas heated or ejected by the central black hole or supernovae	Broad, ~0.1–few Gyr	Outflows, X-ray cavities; operates in centrals too, not just satellites