A wind that carries the ledger of its loops

The Sun loses a million tonnes of gas a second in two quite different winds. The fast wind, at seven or eight hundred kilometres a second, streams from coronal holes — regions where the field lines run straight out into space and never return. It is steady, uniform, and well understood as gas that escapes along field lines that were always open. The slow wind, at three to four hundred kilometres a second, is variable, patchy, and denser, and it fills the space round the Sun’s equatorial belt for most of the solar cycle. Where it comes from has been argued for half a century.

A corona held by its field and opened by its gas located one candidate: the surface, a few solar radii out, where the gas pressure catches up with the falling magnetic pressure again and the closed loops of the streamers give way. If the slow wind is gas that was held in closed loops and released there — at the tops of the streamers, or along the boundaries where closed loops meet open field — it should carry some trace of having been held. It does. The trace is in its composition, and it was recorded long before anyone could fly a spacecraft close enough to watch the release.

Elements sorted by an electron #

Measured in the solar wind at the Earth, the slow wind’s composition differs from the photosphere’s in a peculiar way. Some elements are over-abundant by about a factor of three, relative to oxygen; others are not.

The enhanced elements are not the heavy ones or the light ones, nor the abundant or the rare. They are the ones that are easy to ionise: potassium, sodium, calcium, magnesium, iron, silicon, all with first ionisation potentials below about eight and a half electronvolts. Sulphur and carbon, a little above ten, are partly enhanced. Oxygen, nitrogen, argon and neon, above thirteen, are not; helium, at nearly twenty-five, is depleted. The dividing line falls at about ten electronvolts, the energy of a photon of hydrogen’s Lyman-α line — the most abundant photon in the solar chromosphere — and the coincidence is not one: the line marks the energy below which the chromosphere’s own radiation keeps an element ionised.

The enhancement is relative, and the choice of reference matters. Measured against hydrogen rather than oxygen, the question becomes whether the low-potential elements are enhanced or the high-potential ones depleted, and the answer from the best measurements is mostly the former: iron and magnesium are added to the corona above their photospheric share, rather than oxygen being removed. That distinction is what points to a force that lifts ions, rather than one that holds neutrals back, and it is why the sorting is tied to the magnetic field, which acts only on the charged.

The same pattern, with about the same factor of three to four, is seen in spectra of the corona above active regions and in the closed loops of the quiet Sun. It is much weaker in coronal holes and in the fast wind. That is the first clue: the slow wind’s composition is the composition of closed field.

A temperature at which some elements are ions #

A sorting by ionisation potential needs a place where the potential matters — where the gas is hot enough to ionise some elements and not others — and a force that acts on ions and not on neutral atoms.

That place is the chromosphere, the thin layer above the visible surface where the temperature rises from five thousand to about ten thousand kelvin before the jump to the million-degree corona. In its middle layers Saha’s balance between ionisation and recombination leaves iron, magnesium and silicon almost entirely ionised while oxygen, nitrogen and neon remain neutral. Carbon, intermediate, is partly ionised, which is why it is partly enhanced. Higher up everything is ionised and the opportunity has passed: the separation, if it happens, must happen in the chromosphere’s middle few hundred kilometres.

A magnetic field acts on ions and not on neutrals: ions spiral round field lines and are carried by forces on the plasma, neutrals drift freely through it. So any force exerted by the field, or by waves travelling along it, that pushes plasma upward in the chromosphere pushes the low-potential elements with it and leaves oxygen and neon behind. The leading explanation is the ponderomotive force of Alfvén waves — the average push exerted by waves travelling along the field on the ions that move with them — which is strongest where the waves are reflected, at the steep density gradient at the top of the chromosphere — the same waves whose role in deciding where a field’s structure ends makes them a recurring character in magnetised plasmas from the Sun to the early universe. In closed loops, where waves travel back and forth between the two footpoints and resonate, the force is strongest and acts longest. On open field lines the waves leave and do not return.

The window is not fixed; it moves with the density of the gas.

In a thinner gas recombination is slower and every element is ionised at a lower temperature, so the window in which the low-potential elements are ions and the high-potential ones are not slides cooler. It always exists, because the gap of five or six electronvolts between iron and oxygen is far larger than the shift a factor of a hundred in density produces, but its height in the chromosphere depends on the local density and temperature profile. That is one reason the size of the enhancement differs between loops, coronal holes and the coronae of other stars: the force acts wherever the window lies, and the window lies at different heights in differently structured atmospheres.

A loop’s composition is a clock #

If the sorting happens in closed loops, it should take time, and plasma held in a loop for longer should show more of it.

Spectroscopic observations of active regions as they emerge from below the surface show exactly that. The newest loops have photospheric composition; over a day or two the low-potential elements rise; after several days the loops reach the full enhancement of three to four. In the oldest active regions the enhancement can rise further still. The composition of a loop is a measure of how long its plasma has been confined, the way the darkness of a sunspot is a measure of the field’s share of the pressure — a property of the plasma that reports on the magnetic structure holding it.

Plasma that streams straight out of a coronal hole along open field lines spends only hours in the chromosphere and low corona before leaving, too short for the full sorting, and it arrives in the fast wind with composition close to the photosphere’s. Plasma that has been held in a closed loop for days carries the full enhancement. Released into the wind, it carries that record with it, unchanged, for the four days it takes to reach the Earth, because once all the elements are ionised in the corona nothing sorts them again.

How much of the slow wind was held #

The record can be read quantitatively. The iron-to-oxygen ratio, which spacecraft at the Earth measure every hour, is a mixing ratio between plasma from closed loops and plasma from open field.

Read this way, the slow wind is mostly the contents of closed loops released into space — not entirely, since its composition varies from hour to hour and some streams of it look more like fast wind than like loops, but on average a large majority. The fast wind, by the same reading, has had almost no closed-loop plasma mixed into it. The composition says that the two winds come from different magnetic environments, which is the slow wind’s origin stated in the language of abundances.

Where the closed field is opened #

The question the composition leaves is how plasma in a closed loop gets out. A closed loop is, by definition, a field line with both ends on the Sun, and plasma frozen to it cannot leave without the field line being opened — by reconnection with an open field line nearby, which swaps the loop’s connection so that one end now leads into space. That happens where closed and open field lie side by side: at the boundaries of coronal holes, and at the cusps of the helmet streamers, where the outermost closed loops meet the open field that has pulled away above them. Both are places where plasma beta returns to about one, and where the field can no longer hold the gas against its own pressure.

The Parker Solar Probe, flying closer to the Sun than any spacecraft before it, has found the slow wind full of small structures — magnetic flux ropes, density blobs and abrupt reversals of the field direction, called switchbacks — that look like pieces of closed loops pinched off and released. Solar Orbiter, carrying composition instruments to within a third of an astronomical unit, can match the composition of individual streams to their sources on the Sun’s surface, traced back along the Parker spiral. In the cases examined so far, the slow wind with the strongest enhancement maps back to the edges of active regions and the boundaries of coronal holes, where closed and open field meet.

That is not the only candidate. Some of the slow wind may be gas from the edges of coronal holes themselves, where the open field lines are strongly curved and expand rapidly; such gas flows more slowly for reasons of geometry rather than of origin, and could pick up some enhancement if the field near the footpoints is mixed. The composition alone narrows the choices without closing them: a wind with three times the iron cannot come from a clean coronal hole, but whether it came from a loop that was opened or from the tangled edge of a hole is a question about the last few thousand kilometres of its journey.

A record the Sun’s activity changes #

The sorting is controlled by the magnetic field and the waves on it, so it follows the solar cycle. The butterfly diagram of sunspot latitudes is a magnetic clock, and the wind’s composition keeps the same time. At solar maximum, when active regions are numerous and coronal holes small, the slow wind fills most of the heliosphere and its average enhancement is high; at minimum, when the large polar coronal holes dominate and the polar fields carry the memory of the cycle, the fast wind fills the high latitudes and the slow wind is confined to a band round the equator. The composition of the wind at the Earth, averaged over months, varies with the cycle accordingly, and the long records of solar-wind composition from spacecraft since the 1990s show it.

The same effect is seen in other stars, through the X-ray spectra of their coronae. The most active stars show the opposite — an inverse effect, with low-potential elements depleted — and stars of intermediate activity show little either way. The inversion is not understood in detail, but it fits the picture that the sign and size of the sorting depend on the waves and the field geometry at the base of the corona, which change with a star’s activity. A composition anomaly that began as an oddity in the solar wind has become a diagnostic of how stellar coronae are heated and fed.

The inversion can be pictured as the same clock run backwards.

In the ponderomotive picture the sign of the force depends on where in the chromosphere the waves are reflected relative to the layer where the low-potential elements are ionised: a reflection above the layer pushes ions up into the loop, one below pushes them down out of it. Stars with stronger fields and different wave spectra could put the reflection on the other side, and the sorting runs the other way. That such a simple change in geometry reverses an observable abundance pattern is what makes the composition a sensitive probe of conditions at the base of a corona that cannot be resolved on any star but the Sun.

Why it matters at the Earth #

The slow wind is what most often fills the space between the Sun and the Earth, and its structure — the boundaries between slow and fast streams, the compressed regions where fast wind overtakes slow — decides much of the space weather the Earth experiences between eruptions. A flare’s particles arrive from somewhere else when the field line connecting the Earth to the Sun is rooted far from the flare, and identifying where on the Sun that field line is rooted is the same problem as tracing a slow-wind stream to its source. Composition is the most direct tag. A stream’s iron-to-oxygen ratio and its charge states, frozen in at the temperature of the low corona, together identify the kind of region it left, and so the magnetic connection the Earth is sitting on.

A wind that arrives at the planets with its history #

The composition matters beyond the Sun. The slow wind’s plasma, with its extra iron, magnesium and silicon, is what the planets’ magnetospheres and atmospheres meet most of the time, and its density and speed decide how strongly it shapes a planet that borrows the wind’s field — Venus and Mars, which have no global field of their own, and comets, whose ion tails are made of the wind’s field draped round them. The wind also implants itself in every airless surface it strikes, so the regolith of the Moon and of asteroids holds a record of its composition stretching back billions of years. To read that record the wind’s present composition has to be known separately for each of its regimes, which is why the Genesis spacecraft spent more than two years outside the Earth’s magnetosphere exposing separate collectors to slow wind, fast wind and coronal mass ejections, and returned them for laboratory analysis. The regime-sorted samples showed the same ordering by ionisation potential that the in-flight instruments had found, measured with the precision of a mass spectrometer on the ground — a confirmation of the sorting from atoms that had actually been in the wind.

Rounded abundances and an equilibrium chromosphere #

The figures simplify each part of a chain that is understood only in outline. The abundance enhancements are representative values, rounded; real measurements vary with wind speed, solar cycle and the stream examined, and the enhancement of individual elements has uncertainties of tens of per cent. The ionisation fractions are from Saha’s equation in local thermodynamic equilibrium at a single density, while the chromosphere is far from equilibrium and its ionisation is set by radiation as much as by collisions; oxygen’s ionisation in particular is locked to hydrogen’s by charge exchange, which the figure ignores. The growth of the bias with a loop’s age is a single exponential fitted to a phenomenon that varies from region to region. And the mixing figure assumes two end-member compositions where the real sources form a continuum. The picture — a sorting low down in closed loops, carried unchanged into the wind — survives all of these; the numbers are approximate.

Still open: what lifts the ions #

The ponderomotive force of Alfvén waves accounts for the size of the enhancement, its dependence on the ionisation potential, its growth in closed loops and its weakness in open field, and it predicts the inverse effect in some stars. It is not the only proposal, and it rests on the presence of waves of the right frequencies at the top of the chromosphere, which have been inferred rather than measured. The test it faces is the detailed pattern of the enhancement among the intermediate elements — sulphur, carbon, phosphorus — whose partial enhancement depends on exactly where in the chromosphere the force acts and how strongly. Composition instruments on spacecraft close to the Sun, together with spectra of the same regions from the ground and from orbit, are now measuring that pattern stream by stream. If it matches, the slow wind’s composition becomes a measurement of waves at the base of the corona that cannot be observed directly. If it does not, the threefold excess of iron in the wind that blows past the Earth is a record whose writing is still not understood.