Snow Line

A line drawn in frost

Imagine standing in the disk of gas and dust around the newborn Sun, four and a half billion years ago, and walking slowly outward. Close to the star it is hot — hundreds of kelvin — and water exists only as vapor mixed into the gas. Keep walking. At a certain radius the temperature drops to about 150–170 K, and something abrupt happens: the water vapor freezes, plating itself onto every dust grain in a coat of ice. That radius is the snow line (also called the frost line or ice line), and in the early solar nebula it sat at roughly 2.7 AU — out past Mars, in what is now the asteroid belt.

The snow line is far more than a temperature contour. It is the single most important compositional boundary in a planet-forming disk, because water is, after hydrogen and helium, the most abundant condensible material in the universe. Inside the line, the only things that can be solid are refractory rock and metal — silicates, iron, a little carbon. Outside the line, all of that plus water ice is solid. Because oxygen is so abundant, freezing the water roughly doubles to triples the mass of available solids: the solid surface density Σ_solid jumps by a factor of about 2 to 4 the moment you cross outward. That jump is what builds giant planets.

How the jump builds planets

Planet formation in the core-accretion picture is a competition against time. The gas disk lasts only a few million years before stellar winds and photoevaporation blow it away. To make a gas giant, you must first assemble a solid core of roughly 10 Earth masses fast enough that there is still nebular gas around for it to capture. Once a core reaches that threshold, gas accretion runs away and the planet balloons to hundreds of Earth masses in a geological eyeblink.

The rate at which a core grows depends, to first order, on how much solid material is lying around to sweep up — that is, on Σ_solid. The runaway-growth timescale scales roughly as the inverse of the surface density. So a factor-of-3 jump in Σ_solid at the snow line translates into cores that grow several times faster on the cold side of the line than the warm side. Combine that with the larger feeding zones at greater orbital radius, and the outer disk wins the race decisively: it builds 10 M⊕ cores before the gas disperses, and they become Jupiter, Saturn, Uranus, and Neptune. Inside the snow line, the depleted solid budget never assembles a core large enough to grab gas, so you get only the small, rocky terrestrial planets.

That is the textbook answer to a question that puzzled astronomers for decades: why are all four giant planets in the outer solar system, and all four small rocky planets in the inner solar system? The snow line draws the boundary.

Worked example: locating the line and sizing the jump

Start with the temperature structure. A passively heated, optically thin disk reradiates absorbed starlight, giving a midplane temperature that falls with distance roughly as

T(R) ≈ 280 K × (L / L☉)^(1/4) × (R / 1 AU)^(−1/2)

For a young solar-luminosity star, set T equal to the water condensation temperature. Water ice does not sublimate at a single sharp value; in the low pressures of the nebula it is around 150–170 K. Solving for R at T = 160 K:

160 = 280 × (R / 1 AU)^(−1/2)
(R / 1 AU)^(1/2) = 280 / 160 = 1.75
R = 1.75² ≈ 3.1 AU   (passive disk)

A purely passive disk gives about 3 AU. Add viscous accretion heating in the early, actively accreting phase and the inner disk runs hotter, pushing the line outward; as the accretion rate fades over a few Myr the line creeps back inward. Detailed models that include opacity and accretion luminosity converge on the canonical figure of ~2.7 AU for the time when the giant-planet cores were being assembled. The point to retain: the snow line is a moving target, not a wall, and it sweeps inward across the asteroid belt as the disk cools.

Now size the density jump. In a solar-composition gas, the refractory (rock + metal) condensible mass is a fraction of order 0.3–0.5 percent of the gas. Water, when frozen, adds a comparable or larger condensible mass because of oxygen's abundance. A common bookkeeping:

Inside snow line:  Σ_solid = Σ_rock                  (rock + metal only)
Outside snow line: Σ_solid = Σ_rock + Σ_ice

Σ_ice / Σ_rock ≈ 1 to 3      (depends on assumed C/O, condensation set)
⟹ Σ_solid jumps by ×2 to ×4 across the line

In Hayashi's classic minimum-mass solar nebula (1981), this discontinuity is built in explicitly as a factor of ~4 step in Σ_solid at 2.7 AU, chosen precisely so the model reproduces the masses of the actual planets. The snow line is, in that sense, written directly into the most-used model of the solar system's birth.

A staircase of snow lines

Water is only the first and most consequential condensation front. Every volatile has its own snow line set by its sublimation temperature, and walking outward through the disk you cross them one after another like a frozen staircase:

Volatile	Condensation T	Snow line (solar nebula)	Effect on solids	Body it shapes
Water (H₂O)	~150–170 K	~2.7 AU	×2–4 jump in Σ_solid	giant-planet cores, asteroid belt split
Ammonia (NH₃)	~80–100 K	~5 AU	adds N-bearing ice	Jupiter / Saturn region volatiles
Carbon dioxide (CO₂)	~70 K	~10 AU	further solid boost	Saturn–Uranus ices
Carbon monoxide (CO)	~20–25 K	~30 AU	locks up most carbon	Neptune, Kuiper Belt objects
Nitrogen (N₂)	~20 K	~30–40 AU	traps N₂ ice	Triton, Pluto surface ices
Methane (CH₄)	~30 K	~30 AU	methane-rich ices	outer Kuiper Belt, Pluto

Each front imprints a chemical signature. Bodies that formed beyond the CO snow line carry CO and N₂ ices that those inside it lack — which is why Pluto and Triton have nitrogen and methane on their surfaces while a main-belt asteroid does not. The ratio of carbon to oxygen locked in a forming giant planet's atmosphere depends on which snow lines its building material crossed, a fact now used to "tag" the birth radius of exoplanets from their atmospheric C/O ratios.

Observational status

For most of the twentieth century the snow line was a theorist's construct. That changed with the Atacama Large Millimeter Array (ALMA). In 2013, ALMA detected the CO snow line in the disk around the young star TW Hydrae, not by seeing CO directly but by mapping N₂H⁺ emission — a molecule that is destroyed by gaseous CO and therefore lights up exactly where CO freezes out, at about 30 AU. It was the first direct localization of a snow line in another planetary system.

The water snow line is harder to image because in a normal disk it sits at a few AU, too close to resolve. Nature provided a workaround: the disk around V883 Orionis underwent an accretion outburst that heated it dramatically, pushing its water snow line out to roughly 40 AU. ALMA resolved that swollen line in 2016, capturing a water condensation front in another system for the first time.

Closer to home, the solar system preserves a fossil snow line in the asteroid belt. The inner belt is dominated by dry, silicate S-type asteroids; the outer belt is rich in water-bearing, carbonaceous C-type asteroids — and the transition occurs right around 2.7 AU, marking where the line sat as the planetesimals froze in. The hydrated C-types are widely thought to have delivered much of Earth's water after the planet formed dry inside the line, a connection their D/H isotope ratios support.

Common pitfalls and misconceptions

• "The snow line is a fixed wall." It is a moving, fuzzy front. As the disk cools and the stellar accretion luminosity fades over a few Myr, the water snow line migrates inward by an AU or more. Quoting "2.7 AU" is shorthand for its location during the giant-planet core-building epoch, not a permanent address.
• "Only water has a snow line." Every volatile condenses at its own temperature, producing CO₂, CO, N₂, CH₄, and NH₃ lines at progressively larger radii. The water line is just the most dynamically important because water is so abundant and freezes relatively close in.
• "The snow line is where the disk gets cold enough to be solid." The disk has plenty of solids inside the snow line — rock and metal. What changes at the line is the addition of water ice, not the onset of solids. The relevant quantity is the jump in Σ_solid, not its first appearance.
• "Earth is dry because it formed inside the snow line, so Earth should have no water." Earth did accrete essentially dry, but its water was delivered later by icy/hydrated planetesimals scattered inward from near and beyond the line. The snow line explains the default dryness, not the final inventory.
• "The snow line directly causes planets to form there." It does not place planets; it raises the local solid surface density and may also trap drifting pebbles via a local pressure bump, both of which accelerate core growth. The planet still has to be built by accretion — the line just stacks the odds.
• "Hot Jupiters disprove the snow line." Hot Jupiters orbit far inside any snow line, but they are thought to have formed beyond the line and then migrated inward through the gas disk. Their existence is consistent with — indeed motivated by — outside-the-line formation followed by migration.

Quantitative analysis: why the timescale collapses

To see concretely why crossing the snow line speeds up core growth, take the oligarchic-growth estimate for the runaway-accretion timescale of a protoplanet sweeping up planetesimals. To order of magnitude,

τ_grow  ∝  M^(1/3) / ( Σ_solid · Ω )

where M is the growing core mass, Σ_solid is the local solid surface density, and Ω is the orbital angular frequency. Hold M and R fixed and compare just across the snow line. The orbital frequency Ω barely changes over the narrow radial step, so the timescale ratio is set almost entirely by the density jump:

τ_grow(inside) / τ_grow(outside)  ≈  Σ_solid(outside) / Σ_solid(inside)
                                   ≈  2 to 4

A core just outside the line grows two to four times faster than an identical core just inside it. Over the few-Myr lifetime of the gas disk that is the difference between reaching the ~10 M⊕ runaway threshold in time to capture gas, and falling short. Layer on two reinforcing effects and the contrast sharpens further: (1) icy grains are stickier and survive collisions better than bare silicate, so they coagulate into planetesimals more readily; and (2) the sudden change in opacity and gas pressure at the line can create a local pressure maximum that halts the inward radial drift of mm–cm "pebbles," dumping them just outside the line in a traffic jam. Both pile extra solids onto the outer side of the boundary, on top of the bookkeeping factor of 2–4 from the ice itself.

The net result is the architecture we actually observe: four gas/ice giants whose cores assembled on the cold, solid-rich side of the line, and four small rocky planets starved of solids on the warm side. The same logic, applied to the thousands of exoplanetary systems now known, predicts that giant planets should preferentially appear at or beyond their host's snow line — a trend the data broadly support, once inward migration is folded in.