Coronal Rain

The Sun rains on itself

Stand on the limb of the Sun in the right wavelength and you will watch it rain. Not water, but plasma: clumps of gas at about 10,000 degrees, glowing in hydrogen and calcium light, sliding down invisible magnetic arches toward the surface at the speed of a rifle bullet. The arches are coronal loops, some of them taller than ten Earths stacked end to end, and the cool clumps trace them like beads on a wire. This is coronal rain — one of the most visually arresting and physically revealing phenomena in the solar atmosphere.

The strange part is where the rain comes from. The corona is the Sun's outer atmosphere, and it is searingly hot — one to three million kelvin, hundreds of times hotter than the visible surface beneath it. So the rain is not falling from a cloud of cold gas that drifted in. It is the corona itself, condensing out of its own million-degree heat. Somewhere up on the loop, hot tenuous plasma abruptly loses its ability to stay hot, cools by a factor of a hundred in a few minutes, and crashes back down as cool dense rain. Understanding why a million-degree gas would spontaneously do this is the whole story.

Thermal nonequilibrium: the engine

Coronal rain is the smoking gun of thermal nonequilibrium (TNE). A coronal loop is a magnetic flux tube with both feet rooted in the dense chromosphere and its apex arching up into the corona. Plasma is frozen to the field, so it can only flow along the loop, never across it. The loop's temperature profile is set by a balance between three energy terms: a heating rate E_H, downward thermal conduction along the field, and radiative losses.

The key is where the heating is deposited. If a loop is heated roughly steadily and the heating is concentrated low down, near the footpoints, the energy is delivered to the dense, low-lying plasma. Conduction carries some of it down into the chromosphere, which responds by evaporating fresh material up into the loop. The apex fills with this evaporated plasma but receives little direct heating of its own. There is then no steady temperature the summit can settle into — hence "nonequilibrium." The plasma at the top keeps trying to reach a balance it cannot maintain, and the resolution is catastrophic cooling.

Why cooling runs away

The runaway is a classic thermal instability, first analysed by George Field in 1965 and Eugene Parker in 1953 in slightly different contexts. The radiative loss per unit volume of an optically thin plasma is

L_rad = n_e n_H Λ(T)   [erg cm⁻³ s⁻¹]

where n_e and n_H are the electron and hydrogen number densities and Λ(T) is the optically thin radiative loss function — a tabulated curve set by atomic line emission. The crucial feature of Λ(T) is that, in the range from about 10⁷ K down toward 10⁵ K, it rises as the gas cools (roughly Λ ∝ T^(−1/2) over part of that range). So a small drop in temperature makes the gas radiate more efficiently, which cools it further, which makes it radiate yet more. The instability criterion, in its simplest isobaric form, is that cooling runs away wherever

dΛ/dT < 0   (the loss function increases as temperature falls)

Once a parcel of apex plasma tips into this regime, conduction and heating cannot keep up. The temperature plummets from ~10⁶ K to ~10⁴ K. Because the cooling is nearly isobaric (pressure equalises along the loop much faster than the gas cools), conservation of pressure P = n k_B T means that as T falls by 100×, the density n rises by roughly 100×. The result is a cool, dense condensation — a blob 100 to 1000 times denser than the corona around it — sitting on the loop with nothing to hold it up. Gravity takes over, and it falls.

Falling: slower than you'd expect

Naively, a blob released at a loop apex 50,000 km up should accelerate under the solar surface gravity g_⊙ ≈ 274 m/s² and arrive at the footpoint at the free-fall speed

v_ff = √(2 g_⊙ h)
     = √(2 × 274 m/s² × 5.0×10⁷ m)
     ≈ 1.66×10⁵ m/s
     ≈ 166 km/s        (for a 50,000 km drop, ignoring loop curvature)

But coronal rain is observed to fall at only 30–120 km/s, most often 50–100 km/s — a third to two-thirds of free-fall. The deficit is real and informative. As a dense clump descends, it plows into the plasma below it and compresses it; the gas pressure of the loop leg builds up beneath the blob and pushes back. This "pressure brake" decelerates the rain so that it never reaches free-fall speed. The competition between gravity pulling the dense clump down and the gas-pressure gradient holding it up is the same physics that suspends prominences — coronal rain is just the case where the magnetic geometry offers no resting place, so the balance tips toward falling.

The numbers

Concrete figures sharpen the picture. The table collects the canonical values measured for active-region coronal rain.

Quantity	Typical value	Comparison / note
Coronal (loop) temperature	1–3 × 10⁶ K	~250× the 5772 K photosphere
Condensed rain temperature	~10⁴ K	Chromospheric; emits in Hα, Ca II
Temperature drop	factor ~100	In ~5–20 minutes
Coronal density	~10⁹ cm⁻³	—
Rain blob density	10¹¹–10¹² cm⁻³	100–1000× denser than corona
Fall speed	50–100 km/s	⅓–⅔ of ~166 km/s free-fall
Acceleration observed	~80–100 m/s²	Well below g⊙ = 274 m/s²
Strand width	~100–500 km	Near the 200 km telescope limit
Blob length	~500–2000 km	—
Loop / fall height	10⁴–10⁵ km	Earth diameter ≈ 1.3×10⁴ km
TNE cycle period	~2–8 hours	Heating → evaporation → rain → repeat

The mass involved is not negligible. Integrated over the whole Sun, the rate at which coronal rain drains plasma back to the surface is estimated to be comparable to the total mass of the corona itself over a span of hours, making it a major leg of the chromosphere–corona mass cycle rather than a sideshow.

How we see it

Because the condensed plasma is cool (~10⁴ K), coronal rain shines in chromospheric and transition-region lines, not in the hot EUV channels that show the surrounding corona. That gives observers two complementary views:

• Chromospheric lines. Hα (656.3 nm) and Ca II H & K trace the coolest, densest cores of the rain. Ground-based telescopes with adaptive optics — the Swedish 1-metre Solar Telescope (SST), GST at Big Bear, and the 4-metre Daniel K. Inouye Solar Telescope (DKIST) — resolve individual strands down to ~100–200 km.
• Transition-region and EUV lines. NASA's IRIS (launched 2013) images Si IV (~80,000 K) and Mg II, catching the rain as it cools through transition-region temperatures. SDO/AIA's 304 Å channel (He II, ~50,000 K) shows rain on disk and at the limb; cooler condensations appear dark in absorption in the hot 171 Å and 193 Å channels. Solar Orbiter's EUI has imaged fine-scale "rain showers" at unprecedented resolution from close to the Sun.

A telltale signature in spectra is the appearance of strong redshifts in cool lines low on a loop — material moving toward the footpoint and away from the observer — recurring quasi-periodically as successive TNE cycles produce rain on the same loop bundle. Time-lapse AIA movies of active regions reveal the same arches "raining" over and over through the day.

Where it shows up

• Active-region coronal loops. The classic site. Strongly, footpoint-heated loops above sunspot groups undergo TNE cycles and produce showers of rain along their legs, most spectacular when seen at the limb against the dark sky.
• Flare loops (post-flare rain). After a solar flare, the hot flare arcade cools through coronal temperatures and produces dramatic, large-scale "post-flare loops" of rain — among the brightest and most ordered rain events, draining the energy and mass dumped into the loops by the flare.
• Prominence / filament condensations. The same thermal-instability physics builds the cool material of solar prominences. Where the field has a dip, the condensation is supported and rests as a long-lived prominence; where it does not, it falls as rain. The two are end members of one process.
• Quiescent "long-period" rain. Even outside active regions, long, faint loops can host slow, recurring condensations on multi-hour periods — the gentler, large-scale cousin of vigorous active-region rain.
• Beyond the Sun. The same thermal-instability mechanism is invoked for cool condensations in magnetically structured stellar coronae and is closely related to the runaway cooling that forms multiphase gas in interstellar and intracluster media — coronal rain is the Sun's resolvable laboratory for a universal process.

Common misconceptions and edge cases

• "The rain is cold." It is cool only relative to the corona. At ~10⁴ K it is still hotter than any flame on Earth and roughly twice the temperature of the photosphere it falls toward. "Cool" in solar physics means ~10⁴ K, not room temperature.
• "It falls at free-fall speed." No — the observed 50–100 km/s is well below the ~166 km/s free-fall value. Ignoring the gas-pressure brake leads to overestimating impact energies and misreading the loop dynamics.
• "Coronal rain and prominences are unrelated." They are the same condensation physics in different magnetic geometries. A prominence is essentially suspended rain; rain is essentially an unsupported prominence.
• "It's caused by the gas being heated too much." Subtle but important: it is caused by heating that is steady and footpoint-concentrated. The apex isn't heated enough to stay hot. Spreading the same heating uniformly along the loop would suppress the rain. The distribution of heating, not just its amount, is what matters.
• "Optically thin cooling always applies." The earliest, hottest phase of cooling is optically thin and follows n² Λ(T). But the densest condensed cores can become optically thick in the strongest chromospheric lines, which modifies the late cooling and the line profiles used to measure speeds — a real complication for detailed modelling.