Solar Wind

Parker's 1958 argument: why no static corona is possible

Eugene Parker's 1958 paper begins with a simple question: assume the solar corona is hydrostatic — held up by its own thermal pressure against gravity, with no bulk motion. What does the pressure look like at infinity? For an isothermal corona at temperature T = 10⁶ K with mean particle mass m̄ = 0.6 m_p, the hydrostatic equilibrium equation

dP/dr = −G M_☉ ρ / r²

combined with the ideal-gas law P = ρ k_B T / m̄ integrates to a pressure that does not vanish at infinity but instead asymptotes to a finite, non-zero value:

P(∞) ≠ 0

This is unphysical. The interstellar medium has finite pressure (roughly 10⁻¹³ Pa) but a finite asymptotic coronal pressure that dramatically exceeds it would imply a permanent, static pressure imbalance — gas pushing out against vacuum forever without expanding. The only sensible resolution, Parker argued, is that the corona is not in fact static. It must flow outward.

Parker then derived the steady-state, isothermal radial momentum equation for a smoothly accelerating spherically symmetric flow:

v dv/dr = −(1/ρ) dP/dr − G M_☉ / r²

and showed it has a critical-point structure analogous to the de Laval nozzle of rocketry. The wind starts subsonic deep in the corona, passes through a unique sonic point at about 4 R_☉ where v = c_s, and asymptotes to a supersonic terminal speed of several hundred km/s. The velocity increases with distance because the pressure-gradient term wins over gravity at large r — a property unique to the right kind of flowing solution. Parker's equation has multiple mathematical branches; only the smoothly-accelerating one is physical.

The 1958 paper was rejected by two referees and only published after an editor's intervention. The dominant counter-view at the time held that the corona was a contained, almost-hydrostatic atmosphere; the idea of a continuous wind seemed exotic. The decisive arrival of Mariner 2 data in 1962 settled the argument empirically.

Fast and slow wind: two physically different streams

Once in-situ data accumulated through the 1960s and 1970s it became clear that the solar wind is not a single uniform flow but two distinct populations with different sources and different physical properties.

	Fast solar wind	Slow solar wind
Typical speed at 1 AU	700–800 km/s	300–500 km/s
Typical density at 1 AU	3 protons/cm³	8 protons/cm³
Proton temperature	~ 2 × 10⁵ K	~ 4 × 10⁴ K
Coronal source	Coronal holes (open field)	Streamer belt boundaries
Heavy-element abundance (Fe/O)	~ photospheric	~ 3× photospheric
Variability	Steady, low fluctuations	Highly variable, ICMEs
Mass flux	~ 2 × 10⁸ p/cm²/s	~ 4 × 10⁸ p/cm²/s

The fast wind is the easier piece of the puzzle. Coronal holes — open-field regions, recognised since Skylab (1973) as dark patches in soft X-ray and EUV images — release plasma directly along the magnetic-field lines, with little processing in the closed corona. The fast wind's heavy-element abundances are close to photospheric values (the FIP — first ionization potential — bias is small), suggesting the gas was lifted out of the photosphere and accelerated quickly without lingering in a magnetic trap.

The slow wind is harder. Its enrichment in low-FIP elements (Fe, Mg, Si) of factors 2–5 over photospheric values implies the source plasma sat in closed magnetic loops long enough for gravitational and electromagnetic separation to produce that bias. Most current models invoke "interchange reconnection" between closed loops in the streamer belt and open field lines at the boundary of coronal holes, peeling off plasma in pulses.

From hypothesis to in-situ measurement

The chronology of confirming and refining Parker's prediction:

Year	Mission / instrument	Key finding
1951	Biermann ion-tail observations	Comet tails always anti-solar — implies continuous outflow
1958	Parker theoretical paper	Hydrostatic corona impossible; supersonic wind required
1962	Mariner 2 plasma analyzer	Continuous proton flux at 350–700 km/s confirmed
1973	Skylab S-054 EUV	Coronal holes identified as fast-wind sources
1990–1995	Ulysses out-of-ecliptic	Fast wind dominates solar polar regions at minimum
1995–present	SOHO, ACE, DSCOVR at L1	Continuous monitoring; basis of operational space-weather
2018–present	Parker Solar Probe	In-situ sampling at < 10 R_☉; switchbacks discovered
2020–present	Solar Orbiter	EUV imaging plus in-situ sampling at 0.3 AU; "campfires" linked to wind

Parker Solar Probe, named after the man who predicted what it now measures, made its first perihelion at 35 R_☉ in 2018 and has progressively closed in. In December 2024 it crossed 8.5 R_☉, the closest approach to the Sun any spacecraft has ever made — flying through the solar wind's acceleration region itself.

Worked example: transit time and energy flux at 1 AU

How long does a parcel of average solar wind take to reach Earth from the Sun, and how much energy does it carry per square meter of Earth's leading face?

Take a representative solar-wind speed of v = 450 km/s. Sun-Earth distance is 1 AU = 1.496 × 10⁸ km. Transit time:

t_transit = 1.496 × 10⁸ km / 450 km/s
          = 3.32 × 10⁵ s
          ≈ 3.85 days

The average parcel of solar wind takes about four days to cross the solar system to Earth's orbit. Now energy flux. The solar wind kinetic energy flux at 1 AU is

F_KE = (1/2) ρ v³

With density ρ = 5 protons/cm³ × 1.67 × 10⁻²⁴ g/proton = 8.35 × 10⁻²⁴ g/cm³ = 8.35 × 10⁻²¹ kg/m³ and v = 4.5 × 10⁵ m/s:

F_KE = 0.5 × 8.35 × 10⁻²¹ kg/m³ × (4.5 × 10⁵ m/s)³
     = 0.5 × 8.35 × 10⁻²¹ × 9.11 × 10¹⁶
     = 3.8 × 10⁻⁴ W/m²
     = 0.38 mW/m²

For comparison, the solar electromagnetic-radiation flux at 1 AU (the solar constant) is about 1361 W/m² — roughly 3.5 million times larger. The solar wind is energetically tiny compared to sunlight, but it dominates the magnetospheric coupling because its momentum is concentrated and aligned and its magnetic field reconnects with Earth's.

Total mass-loss rate from the Sun. Assume isotropic outflow at 1 AU with average density 5 cm⁻³ and average speed 450 km/s:

M_dot = 4π R² ρ v
      = 4π × (1.496 × 10¹¹ m)² × 8.35 × 10⁻²¹ kg/m³ × 4.5 × 10⁵ m/s
      = 4π × 2.24 × 10²² × 3.76 × 10⁻¹⁵
      ≈ 1.06 × 10⁹ kg/s

About 10⁹ kg/s, or 3 × 10¹⁶ kg per year. Over the Sun's 4.6-Gyr main-sequence lifetime this is roughly 10⁻⁴ of the Sun's mass — a small fraction by mass but a major contributor to angular-momentum loss because the wind co-rotates out to many solar radii via the Parker-spiral magnetic field.

Variants and extensions

• Stellar winds. All cool stars with hot coronae drive Parker-type winds. Active young stars (T Tauri, M dwarfs) lose 10–1000× as much mass per area as the Sun and produce vastly stronger stellar winds; massive O and B stars instead drive radiation-pressure winds with v_∞ > 1000 km/s and mass-loss rates of 10⁻⁶ M☉/yr.
• Polar plumes. Density enhancements observed in coronal holes near the poles — narrow, radially elongated structures that may be sites of preferential fast-wind acceleration. SOHO and Solar Orbiter EUV imagery shows them clearly; their role in heating remains unclear.
• Co-rotating interaction regions (CIRs). When fast wind from a long-lived coronal hole catches up with slow wind ahead of it, a compression region forms. CIRs are quasi-stationary in a frame co-rotating with the Sun and produce mild, recurrent geomagnetic disturbances at 27-day intervals.
• The Parker-spiral magnetic field. The interplanetary magnetic field is anchored at the rotating Sun and pulled out by the radial wind, winding into a spiral. At 1 AU the field-radial angle is ~45°; at Saturn it is ~85°. The spiral pattern is the geometric backbone of cosmic-ray modulation in the heliosphere.
• Heliospheric current sheet. The boundary between the two polarity sectors of the Parker spiral is a thin sheet of plasma where the magnetic field reverses sign. It warps with the Sun's tilted magnetic axis to form the "ballerina skirt" — a corrugated structure whose crossings at Earth produce mild geomagnetic activity.

Where solar wind shows up

• Geomagnetic activity. Solar-wind density and speed couple to Earth's magnetosphere via the Akasofu coupling function ε ∝ ρ v² × B² × f(θ). Fast streams (CIRs) cause 27-day-recurring storms at K_p ≈ 4–5; CMEs embedded in the wind drive larger storms. ACE and DSCOVR at L1 give 30–60 minutes of warning.
• Comet tails. Ion tails of comets always point anti-solar regardless of comet velocity, because solar-wind protons and the embedded magnetic field carry the cometary plasma away. Hale-Bopp's blue ion tail in 1997, more than 50 million km long, was a textbook example.
• Lunar weathering. Solar-wind protons and helium implant in the top micrometer of the Moon's regolith, producing solar-wind helium-3 (a possible future fusion fuel). Apollo soil samples show ~ 10⁻⁵ atomic fraction of implanted ³He.
• Atmospheric escape on Mars. Without a global magnetic field, Mars's atmosphere is directly exposed to solar wind. NASA's MAVEN orbiter measures pickup-ion escape rates of ~ 10²⁵ ions/s, accumulating to significant atmospheric loss over geological time and partly explaining why Mars is dry.
• Heliospheric structure. The solar wind expands outward to the termination shock at ~ 90 AU and the heliopause at ~ 120 AU, where it meets the local interstellar medium. Voyager 1 and 2 measured the wind directly out to those distances; IBEX maps the boundary indirectly via energetic neutral atoms.

Parker Solar Probe and the acceleration region

Until 2018 the inner edge of in-situ solar-wind measurement was about 0.3 AU (Helios 1 and 2, 1974–86). Parker Solar Probe — protected by a 2.4 m carbon-composite heat shield — has flown through the corona itself. At its closest perihelion (December 2024, 8.5 R_☉ = 0.04 AU) the spacecraft sat in plasma at 1 million K with magnetic-field magnitude up to 1000 nT and proton density up to 10⁴ cm⁻³.

The most surprising discovery of the mission has been switchbacks: short intervals (seconds to minutes) during which the radial component of the magnetic field reverses sign. The field bends back on itself in an S-shape and returns. Switchbacks were unexpected, are ubiquitous near perihelion, and have been variously interpreted as: signatures of interchange reconnection at coronal-hole boundaries; large-amplitude Alfvén waves that have steepened nonlinearly; or remnants of small-scale eruptions in the low corona. The community has not converged on a single explanation, and the question is currently the most active one in solar-wind physics.

Common pitfalls

• Calling the solar wind "the Sun's atmosphere". The atmosphere is the corona; the wind is the supersonic outflow that emerges from it. The two are joined at the sonic surface (around 4 R_☉) but are physically distinct.
• Confusing CMEs with the solar wind. CMEs are discrete eruptions of magnetized plasma that travel through the solar wind. Solar wind is the continuous background; CMEs are storms in it. The wind is always present, even at solar minimum; CMEs come and go.
• Treating the solar wind as energetic compared to sunlight. Solar wind kinetic energy flux at 1 AU is ~ 0.4 mW/m². Sunlight is 1361 W/m². The wind matters because it carries momentum and magnetic field, not because it carries much energy.
• Forgetting helium and minor ions. Solar wind is 95% protons by number, but helium-2+ at ~ 4% carries 16% of the mass and is essential for FIP-bias studies. Minor ions (O⁶⁺, Fe⁹⁺ to Fe¹⁴⁺) probe coronal-source temperatures via charge-state ratios.
• Assuming the speed at 1 AU equals the speed at the Sun. The wind accelerates from rest at the photosphere to its terminal speed by ~ 20 R_☉. PSP's near-Sun measurements show speeds well below 200 km/s at 10 R_☉ — the asymptotic value is reached only well outside the corona.