|Hugh Hudson (2008), Scholarpedia, 3(3):3967.
|revision #190695 [link to/cite this article]
This article briefly introduces solar activity ( Figure 1), by which we understand the many forms of transient behavior patterns of the Sun, especially in its atmosphere, that depend upon magnetism. The source of the surface magnetism in the Sun lies in the convection zone, the layer of the solar interior just below the photosphere. The magnetic fields underlying stellar activity arise from some combination of convective flows and stellar rotation. A deep-seated dynamo mechanism produces the sunspot-scale fields, but in the quiet Sun other effects may play a role as well.
Active-plasma phenomena similar to solar ones, but often much more powerful, appear in all solar-type stars as well as recently-formed stars (T Tauri) in the process of accretion and in a variety of types of double star. A planetary magnetosphere may have similar plasma processes, such as the aurora.
The proximity of the Sun allows us to study the details of rotation and convection at its surface, and to resolve the various forms of magnetic activity with ground-based telescopes and solar space observatories. Even better, we can sometimes actually sample solar material carried out by the solar wind into the range of in-situ observations by spacecraft orbiting the Earth or elsewhere in the heliosphere. In ordinary white light, the solar surface reveals a statistically uniform pattern of granulation outside the magnetic active regions where sunspots appear. Since Galileo's time, the morphology of sunspots has revealed fascinating complexities, starting of course with the (differential) solar rotation itself. The complexities became even more fascinating in the 19th century, when the 11-year solar cycle was recognized, and in the early 20th century when Hale's spectroscopic observations revealed its inherently magnetic nature. The 11-year sunspot cycle then became half of a 22-year Hale cycle because of the alternating hemispheric polarities.
The observed solar magnetic field
Solar magnetism is mainly observed via the Zeeman splitting of the photospheric Fraunhofer lines. With modern observations this provides vector information (via use of the full Stokes parameters) of the emissions, and in principle one can use this information as the basis for an extrapolation into the chromosphere and corona. Valuable supplemental information is now arriving via other techniques, specifically centimeter-wavelength radio spectroscopy, ultraviolet and X-ray image morphology, and direct coronal Zeeman measurements. These measurements generally show quiet-Sun fields to be concentrated in the network boundaries and vertices, with a large amount of the flux in roughly vertical flux tubes having |B| of order 0.1 T. Much of this flux returns to the photosphere on relatively small height scales, but a substantial fraction penetrates into the corona and even into the solar wind if originating in a coronal hole.
Active-region magnetic fields have different properties. Sunspots have even stronger fields, as large as 0.5 T and with much larger surface areas (sunspot umbrae may have areas 104-5 larger than those of the intense flux tubes in the quiet Sun). Accordingly they transmit much more stress locally into the corona, resulting (empirically) in the active-region concentrations of heating seen in X-ray images. The active-region flux concentrations can also be observed to emerge bodily from the interior via buoyant motions, which can be interpreted theoretically in terms of an upward Poynting flux, which transports a part of the solar luminosity up to the photosphere and then through it.
Large-scale solar activity
Stellar magnetism results from the convective and rotational velocity fields working on the ionized medium. These rather simple flow fields produce surprisingly coherent patterns of magnetism, including the solar cycle itself, Maunder's butterfly diagram, and other more exotic orderings. The sunspots appear at middle to low latitudes. As shown by coronal imaging, these complex magnetic fields simplify to become bipolar and radial at the base of the solar wind. In between the photosphere and the solar wind, the coronal magnetic field - although sustaining some currents - can be determined fairly well from a scalar potential function via mapping from the photospheric Zeeman-splitting observations.
On the largest scales the corona evolves slowly, while on the smallest scales there is a continual flickering of activity that involves energy release leading to heating and plasma injection. Although there are cold inclusions (filaments; see below), the bulk of the coronal volume has temperatures of order 1-3 x 106 K. The corona forms a concentric shell of low plasma beta, with its outer boundary defined by the solar-wind flow. Large-area coronal holes occur, as marked by electron temperatures below 106 K and by the inference of open field lines. These are especially prominent at the poles during sunspot minima, but open field lines can occur even within active regions.
Flares and coronal mass ejections
A solar flare consists of a sudden increase in the solar luminosity, together with associated phenomena. By general agreement such a phenomenon results from the unstable release of energy that had been slowly accumulated as magnetic stress in the coronal volume. Observed in soft X-rays, a flaring magnetic loop may have a fast risetime (down to a few sec, but not well characterized observationally yet) and a much slower decay time (e-folding time for temperature decrease of order an hour in extreme cases). At the time of the initial energy release, termed the impulsive phase, a broad array of non-thermal emissions spreads across essentially the whole electromagnetic spectrum. These reflect strong non-thermal particle acceleration, which dominates the flare energy release. During the impulsive phase a large mass of new material from the lower atmosphere is injected into coronal magnetic loop structures. The gradual phase consists of the processes, notably soft X-ray emission, involved with the cooling of this material. Figure 3 shows a solar flare observed by the Hinode spacecraft, which carries the first high-resolution solar telescope in space. This image is in the Fraunhofer G-band, which forms mainly in the photosphere. Accordingly one can see faculae and network signatures, as well as the large sunspot group. The elongated G-band brightenings are the flare ribbons, which mark the footpoints of coronal magnetic flux tubes containing hot, dense plasma visible most easily in soft (few keV) X-radiation.
The movie to the right shows the full time development of a powerful impulsive flare of January 20, 2005, as observed in the EUV by TRACE and in soft (red) and hard (blue) X-rays by RHESSI. The total elapsed time is 1.5 hours, and the EUV image quality fluctuates. In the movie sequence one can see the formation of ribbons, as in the example of Figure 3, together with hints of the presence of coronal magnetic loops connecting them, and filled with hot plasma as the result of the flare. The solar cosmic rays and high-energy neutrons produced by this event penetrated to ground level on Earth, a rare phenomenon associated with the most energetic events.
A coronal mass ejection (CME) typically accompanies the most energetic flare events.Figure 4 (left) shows this in a composite image that incorporates both a coronagraph view (scattered light from above the limb) and a direct view (emission at high temperatures) of the corona. The flare and CME phenomena are to some extent independent, in the sense that weaker flares tend not to have CMEs, and that large-scale CMEs can take place with only weak flare effects. The CME has the appearance of expanding magnetic flux tubes that entrain plasma and eject it into the solar wind. The expanded flux tubes can achieve large enough radial distances that they may be considered open and support solar-wind flow. Further extensive particle acceleration may result from large-scale collisionless shock waves associated with the ejecta (see below).
Filaments are elongated coronal features composed of relatively dense material at chromospheric temperatures. They are identified with quiescent prominences at the limb, against the disk they are seen in absorption, and on the limb in emission. They reside in filament channels, which consist of elongated, nearly horizontal, and slightly twisted flux tubes. The filament channels and the filaments they support have many interesting properties; they can form in the quiet Sun even at high heliographic latitudes. They may erupt, and this often leads to a CME and to an associated flare-like structure of enormous scale (of order a solar radius) best visible in soft X-ray images or in chromospheric lines. The classical CME morphology has a three-part structure: a bright front, a dark cavity (identifiable with the pre-eruption filament channel), and some filament material embedded in the cavity.
In addition to flares and CMEs, soft X-ray observations of the solar corona reveal the existence of jets on large as well as small scales. Figure 5 shows an excellent example. The motions in the jets appear to be parallel flows of plasma along the magnetic field, with apparent speeds of order 103 km/s. The association of X-ray jets with meter-wave type III radio bursts, which are due to weakly relativistic electrons that may be subsequently observed far out in the solar wind, establishes that in many cases these jet fields mark open field lines. Invariably a compact flaring loop structure appears at the base of a jet, strongly suggesting an interaction between closed and open magnetic field systems involving magnetic flux transfer across the separatrix between the two systems. Figure 1 shows another excellent example of a jet, to be seen in the NW (upper right) quadrant of the first image.
Small-scale solar activity
Flares, microflares, nanoflares
Flare-like activity extends to smaller scales and to lower energies. The weakest events - nominally of order 10-6 of the energy of a major flare - are the microflares. Parker proposed the term nanoflare for still weaker events hypothetically associated with the formation of current sheets and their dissipation. The sticking and slipping of the coronal plasma between these temporary static equilibria, the nanoflares, would then constitute the heating mechanism involved with the extraction of magnetic free energy in the coronal magnetic field stressed by random footpoint motions. The nanoflares would be essentially too numerous to recognize individually, and perhaps not surprisingly no unambiguous evidence for their existence has yet appeared. The microflares, however, are easy to observe by a variety of techniques because of their isolated spatial and temporal distributions.Figure 6 shows their spatial pattern for a five-year interval of time, establishing their strong association with the sunspot regions. The occurrence distribution function for solar flares and microflares, and for stellar flares, tends to follow a flat power law in peak flux F, such as F-1.8. Its flatness implies that the most energetic events dominate the total flux of all flares taken together, rather than the more numerous smaller events.
X-ray bright points
The last image in the sequence of Figure 1 (the rightmost one) shows a solar-minimum corona. The obvious X-ray emission comes from only a few X-ray bright points, phenomena originally discovered with the Skylab soft X-ray telescopes and their rocket-borne predecessors. The bright points tend to have a uniform and time-invariant distribution across the solar surface, as expected from magnetic fields associated with the surface convective flows (the supergranulation of the quiet photosphere) rather than with the active-region fields responsible for sunspot fields.
Activity in the lower atmosphere
In addition to the flare-like phenomena, small-scale activity takes other forms. In the lower atmosphere the jet pattern persists, though "true" flares appear to require coronal conditions. The chromosphere contains a great deal of dynamic activity that is being sorted out by high-resolution observations; the topics include spicules as well as jets and may not involve the high temperatures (above a few 106 K) that characterize flares. Many names are associated with this small-scale activity: jets, turbulent events, network flares, blinkers, explosive events, X-ray bright points, Ellermann bombs, evanescent active regions, etc.
Finally, at the very base of the solar atmosphere and in fact embedded into the photosphere, one finds the G-band bright points and the faculae. These are bright elements resulting from the hot walls of granulation adjacent to intense magnetic flux tubes. This effect results from the reduced opacity created by the substitution of magnetic pressure for gas pressure within the flux tube. There is a tight relationship between photospheric faculae and chromospheric plage, but the hot-wall model does not comment on this and the actual physics remains unclear.
The excess luminosity of faculae alters the total solar irradiance subtly but detectably, as do the sunspots. In more extreme stars this effect can be much larger, and interestingly can reverse the phase of the cyclic brightness variations relative to the solar pattern, where sunspot area maximum corresponds somewhat non-intuitively to the phase of maximum brightness. At solar maximum the Sun is some 0.1% brighter than its solar-minimum level. In more active stars the magnetic activity may have substantial impacts on basic stellar parameters, such as the luminosity, and probably therefore have substantial effects on the stellar structure itself.
Particle acceleration, magnetic reconnection, and energetics
Particle acceleration forms an integral part of solar magnetic activity, in the sense that the corona exhibits the sudden transient effects (flares, CMEs, jets) described above. In most cases the release of energy does not involve heating as such, but rather particle acceleration, in which the distribution function of the plasma becomes highly anisotropic and non-Maxwellian. The movie to the right (click to activate) shows the results of strong particle acceleration by a CME, probably associated with the collisionless shock wave driven ahead of it into the solar wind. In general the acceleration of high-energy particles dominates flare energetics, even on the small scales of the microflares.
The flare component of a major event (i.e., its electromagnetic radiations) and the CME component (its mechanical effects) can each consume as much as 1032 ergs in extreme cases. The relatively low Alfvén speed in the photosphere precludes any sudden injection of energy from this reservoir, and so in general terms we know that the energy of the event must come from a slow accumulation in the coronal magnetic field, followed by a sudden release and partition into the observed forms. We are sure that magnetic reconnection describes some of the important effects, we do not presently understand the physics of the instability that results in the flare and/or CME. Note that the coronal plasma has a low plasma beta, meaning that non-magnetic forces such as gravity cannot play a significant role in the energetics.
Generally solar-type stars have the combination of rotation and convection necessary for the existence of complex surface magnetic fields, and hence stellar activity. This stellar magnetism then leads to solar-type activity, including starspots, stellar flares, and presumably other analogous and also unique forms of energy release. Related activity can also take place in various binary stellar systems and during the early life of newly formed stars; during the T Tauri phase there may be rapid rotation and an accretion disk, and activity occurs at levels much higher than the current solar level. The occurrence of related forms of magnetic activity across a broad spectrum of astronomical objects complements studies of solar activity, for which it is impossible to vary the rotation or gravity, for example, as a means of empirical understanding of the dynamo mechanism.Figure 7 shows the generality of the correlation between activity, as measured by X-rays, and magnetic flux.
Weaker forms of plasma activity occur in planetary magnetospheres. The terrestrial aurora and its linkage with the magnetosphere and its geomagnetic tail provide several often-discussed analogies to flares and CMEs. The drivers of geomagnetic activity include the magnetopause and the ionosphere. Note that these structures do not have strict analogs in the solar case. More importantly than the analogies, though, space plasmas provide a laboratory for the in-situ study of important plasma-physics concepts involving conditions (such as collisionless shock waves) that are difficult to reproduce in the laboratory. The microphysics is of course completely impossible to study in solar or stellar environments by remote-sensing techniques, but not so to spacecraft capable of measuring particles and fields within these space plasmas directly down to the Debye scale. In the domain of space plasma physics, one can in principle obtain full measurements of the particle distribution function and the 3D structure of the plasma.
- Hudson, H., and Ryan, J., High-Energy Particles in Solar Flares, Ann. Revs. Astron. Astrophys. 33, 239 (1995)
- Kahler, S. W., Solar Flares and Coronal Mass Ejections, Ann. Revs. Astron. Astrophys. 30, 113 (1992)
- Parks, G. K., The Physics of Space Plasmas, Perseus, 1991
- Pevtsov, A. A., Fisher, G. H., Acton, L. W., Longcope, D. W., Johns-Krull, C. M.., Kankelborg, C. C., and Metcalf, T. R., Astrophysical Journal 598, 1387 (2003)
- Schrijver, C. J., and Zwaan, C., Solar and Stellar Magnetic Activity, Cambridge University Press, 2000
- Tandberg-Hanssen, E. and Emslie, A. G., The Physics of Solar Flares, Cambridge University Press, 1988
- Axel Brandenburg (2007) Hydromagnetic dynamo theory. Scholarpedia, 2(3):2309.
- Brigitte Schmieder and Etienne Pariat (2007) Magnetic flux emergence. Scholarpedia, 2(12):4335.
- Guenther Ruediger (2008) Solar dynamo. Scholarpedia, 3(1):3444.
- Wolfgang Schmidt (2008) Solar telescopes. Scholarpedia, 3(4):4333.
- Philip Holmes and Eric T. Shea-Brown (2006) Stability. Scholarpedia, 1(10):1838.