Solar Spicules

A million fires at the limb

Photograph the edge of the Sun through a narrow-band Hα filter and the limb does not look smooth. It looks like a burning meadow — thousands of bright, thin, near-vertical threads pointing outward, each only a fraction of an arcsecond wide, each lasting a few minutes before fading. This is solar granulation's wilder upstairs cousin: the spicule layer of the chromosphere.

Father Angelo Secchi, observing the 1877 solar eclipse from the Collegio Romano in Rome, was the first to describe what he saw at the limb: a "sea of flames", short-lived, in vigorous motion. Father Pietro Secchi's spectrohelioscope and later George Hale's spectroheliograph let astronomers freeze single moments of the chromosphere across the disk. The English word "spicule" — Latin spīculum, a small pointed dart — was coined by Walter Roberts in 1945, and it stuck because the threads really do look like darts thrown from the surface.

Modern counts put the population at about 10⁶ spicules covering the Sun at any moment — comparable to the number of granules but much smaller in width. Each spicule occupies a column roughly 200–500 km across (some Type II ones are 150 km or less), rises 5,000–10,000 km into the corona, and dies after about five minutes. That means an average patch of the chromosphere is being criss-crossed by several spicule launches every minute.

Two populations: classic and ballistic

Spicules were originally treated as a single phenomenon, but Hinode's high-cadence Solar Optical Telescope data in 2007 — analysed by Bart De Pontieu and colleagues — revealed two physically distinct populations.

	Type I (classic)	Type II (fast/ballistic)
Typical width	300–500 km	150–250 km
Typical length	5,000–9,000 km	3,000–7,000 km
Apparent ascent speed	15–30 km/s	50–150 km/s
Lifetime	3–7 min	50–150 s
End state	Falls back ballistically	Fades after heating
Sky distribution	Quiet Sun and active regions	Coronal-hole boundaries and network
Identified by	Secchi 1877; Roberts 1945	De Pontieu et al., Hinode, 2007

Type I spicules look like ordinary plasma being lifted up and pulled back by gravity — a ballistic projectile in the chromospheric atmosphere. Type II spicules are the unexpected discovery. They are roughly 5× faster, half as wide, and crucially do not fall back. Instead they fade from chromospheric to coronal temperatures in a few seconds — meaning their material is being added to the hot corona rather than recycled into the chromosphere. They cluster preferentially around the boundaries of magnetic network and at the edges of coronal holes.

The drivers are probably distinct. Type I spicules look consistent with magnetoacoustic shock propagation: the photosphere oscillates on a 5-minute period (p-modes), shocks form in the upper chromosphere, plasma is lifted, then falls. Type II spicules look consistent with magnetic reconnection between emerging flux tubes and pre-existing field, ejecting plasma fast enough to escape and to be heated.

Why coronal-heating physicists care

The corona reaches 1–3 million K while the photosphere just beneath it sits at 5,800 K. Where the energy comes from to maintain this temperature gradient against radiative losses (~10⁷ erg cm⁻² s⁻¹ in the quiet Sun) is the long-running "coronal heating problem". Two broad classes of mechanism are candidates: wave heating (Alfvén or compressive waves generated below the photosphere propagate up, become non-linear, and dissipate in the corona) and reconnection heating (countless small reconnection events — nanoflares — release stored magnetic energy directly into thermal energy).

Spicules sit at the intersection of both. The fast Type II ones carry Alfvénic transverse oscillations of amplitude ~20–30 km/s, observed by Solar Dynamics Observatory and IRIS — sufficient to deliver an energy flux of ~ 4 × 10⁶ erg cm⁻² s⁻¹ if those waves dissipate higher up. They also originate at magnetic-flux boundaries, where reconnection between flux tubes is a natural source. The fading of Type II spicules at high temperatures is evidence that energy is being thermalised in the column they have just lifted.

No single mechanism heats the entire corona. But spicules — particularly Type II — have moved from "interesting limb feature" to "leading candidate for at least part of the heating budget" over the last two decades.

Worked example: total mass flux carried by spicules

Estimate the total upward mass flux carried by Type II spicules and compare it to the solar wind mass-loss rate.

Take a representative Type II spicule. Cross-section radius r = 100 km = 10⁵ m, so area A = π r² ≈ 3.1 × 10¹⁰ m². Ascent speed v = 100 km/s = 10⁵ m/s. Chromospheric density at the launch point: n_e ~ 10¹⁰ cm⁻³ = 10¹⁶ m⁻³, mostly hydrogen, so mass density ρ ~ 1.67 × 10⁻²⁷ kg × 10¹⁶ m⁻³ = 1.67 × 10⁻¹¹ kg/m³.

Mass flux per spicule = ρ × v × A
                       = 1.67 × 10⁻¹¹ × 10⁵ × 3.1 × 10¹⁰
                       = 5.2 × 10⁴ kg/s

Now total: take a fraction f = 0.1 of the 10⁶ spicules visible at any moment as Type II (others are slow Type I). That gives 10⁵ Type II spicules globally:

Total mass flux = 10⁵ × 5.2 × 10⁴ kg/s
                = 5.2 × 10⁹ kg/s
                ≈ 4× solar wind mass-loss

Most of this mass falls back into the chromosphere or remains there. But only ~25% needs to escape onto open field lines to entirely supply the solar wind. The numbers are reasonable for the conjecture that spicules are a major source of solar-wind mass — without being conclusive proof, because how much fraction escapes is exactly the unsolved problem.

Energy flux: take the same spicule with kinetic energy (1/2) ρ v² ≈ 8.4 × 10⁻² J/m³, multiplied by ascent speed and converted to surface flux. Over the full chromospheric surface, this contributes of order 10⁶ erg cm⁻² s⁻¹ — about 10% of what the corona radiates, with another order of magnitude potentially carried by the Alfvénic transverse motions on top of the bulk velocity. The accounting is uncertain but not wildly off.

Observing history

Year	Instrument / observer	Key finding
1877	Father Angelo Secchi (Rome)	"Sea of flames" at solar limb in Hα
1945	Walter Roberts	Coins term "spicule"
1968	Beckers (review)	Quantitative widths, lengths, lifetimes
2002	Swedish Solar Telescope first light	0.1″ ground-based imagery; spicule structure resolved
2006	Hinode launched	High-cadence chromospheric movies
2007	De Pontieu et al.	Type II spicules identified in Hinode
2013	IRIS launched	UV transition-region spectra resolve spicule heating
2019	DKIST first light	25 km resolution, finer than narrowest spicules
2020	Solar Orbiter EUI	"Campfire" features may be spicule-related

Variants and related features

• Macrospicules. Larger cousins seen in EUV light (He II 304 Å in particular) at coronal-hole boundaries. They reach 40,000 km and last 15–20 minutes. Macrospicules carry more mass per event but are far less numerous than Type II spicules.
• Coronal jets. Distinct phenomenon at greater scale — collimated outflows from polar coronal holes seen as bright lanes in SOHO/EIT and SDO/AIA imagery. Driven by reconnection of emerging bipoles with overlying coronal-hole field. Sometimes called "blowout jets" when more energetic.
• Mottles, fibrils. The internetwork chromosphere — quieter regions away from network — shows similar fine-scale jets called mottles or rosette fibrils. They are smaller, slower and more horizontally aligned than typical spicules; the distinction is one of geometry as much as physics.
• RBEs (rapid blueshifted excursions). The on-disk signature of Type II spicules — sudden blueshift of the Hα or Ca II line for ~50–100 s, indicating mass moving upward at the line of sight. RBEs have been crucial for confirming that Type II spicules are real and not limb projection effects.
• Solar tornadoes. Larger rotating columns of plasma at the chromosphere–corona interface, lasting hours. Distinct from spicules but possibly linked via underlying magnetic-flux footpoint motions.

Where spicules show up

• Eclipse imagery. Total-eclipse photographs of the chromospheric flash spectrum (the bright pink ring around the lunar disk seconds before second contact) include the spicule layer. The fuzzy texture of that ring is mostly spicules.
• EUV movies from SDO/AIA at 304 Å. The "boiling pink Sun" texture is mostly spicule activity at the network boundaries.
• Type II spicules on dim coronal-hole boundaries. Believed to feed the slow component of the solar wind.
• JWST stellar-activity context. When JWST tries to subtract stellar contamination from exoplanet atmospheric spectra, spicule-like jets on the target star contribute time-variable line-emission noise.
• Coronal-heating budget accounting. Spicules feature in every modern review of the coronal heating problem as one of three or four main candidate mechanisms.

Common pitfalls

• Calling all spicules the same. Type I and Type II spicules are physically distinct populations with different drivers, lifetimes, end-states. Treating them as one will give wrong predictions for any coronal-heating model.
• Confusing spicules with prominences. Prominences are huge magnetic structures, 100,000+ km long, suspended in the corona. Spicules are 10,000 km or less. Prominences last days; spicules last minutes.
• Imagining spicules are "flames". They are not combustion. They are plasma jets — magnetized gas pushed along magnetic-field lines by reconnection or wave forcing.
• Forgetting they are everywhere. Photographs of the disk usually erase them in averaging; narrow-band Hα cinema is needed to see the population at any moment. The smooth chromosphere is an illusion.
• Overstating their solar-wind role. Spicules may supply much of the slow wind, but most spicule mass returns to the chromosphere; only a small leak feeds open field lines. The mass-loss rate is set by that small leak, not by total spicule activity.