Ann. Geophys.

, 35, 805816, 2017

https://doi.org/10.5194/angeo-35-805-2017
Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Our dynamic sun: 2017 Hannes Alfvn Medal lecture at the EGU
Eric Priest1,*
1 Mathematics Institute, St Andrews University, St Andrews KY16 9SS, UK
* Invited contribution by Eric Priest, recipient of the EGU Hannes Alfvn Medal for Scientists 2017

Correspondence to: Eric Priest ([email protected])

Received: 15 May 2017 Accepted: 12 June 2017 Published: 14 July 2017

Abstract. This lecture summarises how our understanding 1 Introduction

of many aspects of the Sun has been revolutionised over the
past few years by new observations and models. Much of the
Thank you most warmly for the award of the Alfvn Medal,
dynamic behaviour of the Sun is driven by the magnetic field
which gives me great pleasure. In addition to a feeling of
since, in the outer atmosphere, it represents the largest source
surprise, I am also deeply humbled, because Alfvn was one
of energy by far.
of the greats in the field and one of my heroes as a young
The interior of the Sun possesses a strong shear layer at the
researcher (Fig. 1). We need dedicated scientists who fol-
base of the convection zone, where sunspot magnetic fields
low a specialised topic with tenacity and who complement
are generated. A small-scale dynamo may also be operating
one another in our amazingly diverse field; but we also need
near the surface of the Sun, generating magnetic fields that
time and space for creativity in this busy world, especially
thread the lowest layer of the solar atmosphere, the turbulent
for those princes of creativity, mavericks like Alfvn who can
photosphere. Above the photosphere lies the highly dynamic
view our field widely and blaze trails in new and unexpected
fine-scale chromosphere, and beyond that is the rare corona
directions.
at high temperatures exceeding 1 million degrees K. Possi-
The Sun, an object of worship for early civilisations, is
ble magnetic mechanisms for heating the corona and driving
the main source of light and life on Earth and of our space
the solar wind (two intriguing and unsolved puzzles) are de-
weather with many subtle effects on our environment. It is
scribed.
also a key for astronomy, since many of the fundamental cos-
Other puzzles include the structure of giant flux ropes,
mic plasma processes taking place in stars, galaxies and ac-
known as prominences, which have complex fine structure.
cretion discs can be viewed in much greater detail on the Sun.
Occasionally, they erupt and produce huge ejections of mass
However, solar physics is at present in a vibrant state since
and magnetic fields (coronal mass ejections), which can dis-
many of the basic properties of the Sun remain a mystery.
rupt the space environment of the Earth. When such erup-
Although we have made huge progress on them over the past
tions originate in active regions around sunspots, they are
10 years, definitive answers have not yet been given about
also associated with solar flares, in which magnetic energy
how the magnetic field is generated, how the dynamic fine-
is converted to kinetic energy, heat and fast-particle energy.
scale structure of the atmosphere is created, how the corona
A new theory will be presented for the origin of the twist that
is heated and the solar wind accelerated, how prominences
is observed in erupting prominences and for the nature of re-
are formed and structured, how eruptions of coronal mass
connection in the rise phase of an eruptive flare or coronal
ejections are initiated and how reconnection converts energy
mass ejection.
in solar flares.
Keywords. Solar physics astrophysics astronomy (corona These puzzling features are caused by the magnetic field.
and transition region) The Sun is plasma rather than normal gas, so it is coupled in
an intimate and subtle way to the magnetic field. This inter-
action is described by the equations of magnetohydrodynam-
ics (or MHD). To me these are the most beautiful equations
I know and are continually revealing new properties. They

Published by Copernicus Publications on behalf of the European Geosciences Union.

806 E. Priest: Alfvn Medal lecture

a vacuum but is pervaded instead by plasma, which could

carry currents that generate a galactic magnetic field. He
was trained as an electrical engineer and obtained a PhD
from Uppsala in ultra-short electromagnetic waves (1934).
He taught in Uppsala and Stockholm and was appointed a
professor at the Royal Institute of Technology in Stockholm
in 1940. Later, in 1967, he moved to San Diego, Califor-
nia. He was awarded the 1970 Nobel Prize in physics for
his contributions and fundamental discoveries in magneto-
hydrodynamics, and their fruitful applications to different ar-
eas of plasma physics.
The history of ideas is often fascinating. In particular,
Figure 1. Images of Hannes Alfvn as a young man (age 34) and Southwood (2015) has reviewed the history of our under-
older. standing of the aurora. At the beginning of the last century,
Kristian Birkeland, in a ground-breaking move, painstak-
consist of equations of induction, mass continuity, motion, ingly set up polar stations for auroral observations and de-
perfect gas and energy, which take the following form in the duced that the aurora are too high (100 km) to be atmospheric
corona: phenomena. He suggested that they are produced by horizon-
tal electric currents flowing in the ionosphere and closed by
B currents that flow in on one side along the magnetic field and
= (v B) + 2 B, (1)
t out on the other (Birkeland, 1908). These are now known as
d Birkeland currents, but they were only confirmed in 1967.
+ v = 0, (2)
dt Birkeland thought that the charged particles came directly
dv from the Sun. However, for half a century, the idea of field-
= p + j B + F , (3)
dt aligned currents was ignored by most solar terrestrial scien-
!
kB Re tists, who were split into two antagonistic schools, one Scan-
p= T = T , (4) dinavian and the other British. The antagonism and ill feel-
m e
ing between these schools was in my view highly regrettable,
 
d p since it would have been better to listen respectfully to each
= q Lr + j 2 / + FH , (5)
1 dt other by trying to understand the other point of view and re-
main on good terms. The dominant British school was led
where F includes extra forces, q is the heat flux vec- by Chapman, who produced a different theory of geomag-
tor, Lr is the optically thin radiative loss function and netic storms (Chapman and Ferraro, 1930) based on the idea
FH =  + Hv + Hc is a heating term. These equations are, of a closed magnetosphere separated from the interplanetary
in general, coupled and serve to determine the primary vari- medium by a magnetopause. Alfvn, however, doggedly kept
ables, the plasma velocity (v), the magnetic field (B), the Birkelands ideas alive by suggesting that a magnetised parti-
plasma pressure (p), the plasma density () and the plasma cle stream simply flows onto the Earths magnetic field and is
temperature (T ). In addition, the secondary variables, the driven by a voltage that comes from the Earths rotation. Nei-
electric current (j ) and electric field (E), are given explic- ther picture was correct, but both had elements of truth. They
itly in terms of the velocity (v) and magnetic field (B) by were ultimately unified when Dungey (1961) produced his
open-field model of the magnetosphere with the solar wind
j = B/,
driving southward-directed magnetic field lines to reconnect
E = v B + j /, at the nose of the magnetosphere and also in its tail. It is this
reconnection process that drives field-aligned currents in on
while B is subject to the condition
the dawn side and out on the dusk side.
B = 0. Alfvn wrote a highly insightful book in 1950 entitled
Cosmical Electrodynamics with introductory chapters on
This last equation plays the role of an initial condition for the charged particle motion, electrical discharges and Alfvn
time-dependent Eqs. (1)(5), which form a complete set, rep- waves propagating along a magnetic field, followed by ap-
resenting nine equations for nine variables. The divergence of plication chapters on solar physics, magnetic storms and au-
Eq. (1) shows that if B vanishes initially, then it continues rora, and cosmic rays. I recently reread the book and was
to vanish for all time. struck by his many imaginative ideas based on the limited
Alfvn (19081995) was an early pioneer in proposing observations of the time, but also by how much the newer ob-
MHD ideas, such as frozen flux and, in 1942, MHD waves. servations have revolutionised our understanding since then
In 1937 he suggested that the universe is not filled with and made many of Alfvns ideas inapplicable. For exam-

Ann. Geophys., 35, 805816, 2017 www.ann-geophys.net/35/805/2017/

E. Priest: Alfvn Medal lecture 807

ple, he thought that the core of the Sun possesses a uniform Pole 33.1 days
1.0
magnetic field and is surrounded by a global bipolar field 28.9 days
that produces a weak field of 20 G at the solar surface. In his
0.8
mind, the solar interior possesses two turbulent regions, the
first near the surface only a few granules thick and the second
0.6
in the energy-generating core; the core turbulence generated 25.7 days
Alfvn waves in the form of very strong flux rings lying in
0.4
lines of latitude which propagate out along the bipolar field
and create a pair of sunspots when a ring reaches the solar 0.2
surface.
We now know that the Sun is a ball of plasma held together Equator
0.0
by gravity with a radius of 700 Mm (i.e., 700 million m) com- 0.0 0.2 0.4 0.6 0.8 1.0
pared with the Earths radius of 6 Mm. Its interior possesses
a core at which the energy is released by thermonuclear re- Figure 2. A meridional cut through the solar interior, showing
the rotation rate deduced by solar seismology techniques. Rotation
actions and diffuses very slowly outwards. In the outer one-
times are indicated on three thick curves. There are shear layers at
third of the interior the temperature gradient is large enough
the solar surface and at the tachocline (after Schou et al., 1998).
to drive convective instability, and this produces a turbulent
convection zone. The atmosphere consists of three parts: the
thin photosphere with a temperature of 6000 K, from which dition it was expected that the magnetic field would be gen-
comes most of the light we see. Above it lies the warmer and erated throughout the convection zone.
rarer chromosphere with a temperature of about 10 000 K. A huge surprise, therefore, was the discovery from solar
Beyond that the outer atmosphere or corona stretches out to seismology that, throughout the convection zone, the rota-
fill the heliosphere. tion is instead constant on cones (Duvall and Harvey, 1984;
In this lecture I plan to give you some highlights of the Duvall et al., 1984). Also, at the base of the convection zone
progress made since Alfvns Nobel Prize in 1970 that have there is an intense shear layer, called the tachocline, which
come from ground-based and space observations and also is now thought to be the location of the main dynamo where
from theory. These will include accounts of the photosphere, the magnetic field of sunspots and active regions is generated
chromosphere, prominences, coronal heating, solar wind and (Fig. 2). Near the solar surface, one can also see a shear layer,
finally solar flares and coronal mass ejections. Each of these which may possibly be the location of a small-scale dynamo
are huge subjects which have spawned many books and re- responsible for the small-scale magnetic flux that appears at
views, so my discussion will necessarily be simple and se- the solar surface as ephemeral regions, network flux and in-
lective. ner network flux.

3 The low atmosphere

2 The solar interior
3.1 The photosphere
In 1970, the supposed internal temperature T (r) of the Sun
was based purely on theory. The magnetic field of sunspots The photosphere has a temperature of 6000 K and is covered
was generated by a dynamo and rose by magnetic buoyancy. with turbulent convection cells on two main scales. Gran-
Now there are magnificent computational experiments for ulation consists of tiny cells with a typical size of 1 Mm,
the details of how the magnetic flux tubes rise through the whereas supergranulation is much larger with scales of 15
solar surface and interact with the overlying magnetic field 30 Mm.
(e.g., Archontis et al., 2014; Archontis and Hansteen, 2014; Maps of the line-of-sight magnetic field reveal bipolar ac-
Martnez-Sykora et al., 2015; Hansteen et al., 2017). tive regions around sunspots forming two bands, one north
Now we also know that the Sun oscillates in many differ- of the equator and the other south (Fig. 3). In 1970, Alfvn
ent normal modes of vibration. Several million modes have thought there was a weak large-scale bipolar magnetic field
been discovered and form the basis for the new field of so- outside active regions; but by the 1990s, very strong small-
lar seismology, from which the internal temperature has been scale magnetic fields had been discovered at the edges and
deduced (Christensen-Dalsgaard et al., 1985). The internal junctions of supergranules, forming a so-called network.
rotation of the Sun has been deduced and compared with the Now, if we map the photospheric magnetic field above
surface rotation pattern in which the surface equatorial re- a threshold of 500 G, the network flux shows up clearly at
gions rotate faster than the polar ones. It had been thought the edges of supergranules. However, if the threshold is low-
that the internal rotation would be constant on cylinders, with ered to 25 G, far more magnetic flux shows up and tiny ele-
the outer cylinders rotating faster than the inner ones; in ad- ments of a strong magnetic field are revealed at the edges of

www.ann-geophys.net/35/805/2017/ Ann. Geophys., 35, 805816, 2017

808 E. Priest: Alfvn Medal lecture

(a) (b)

Figure 4. Line-of-sight magnetic field from (b) the IMAX instru-

ment on the Sunrise balloon and (a) the same data at the resolution
of the Hinode mission. The bipolar fragments (black and white next
Figure 3. A typical map of the line-of-sight magnetic field in the so- to one another) represent emerging flux if they are separating and
lar photosphere. The bright regions represent the magnetic field that cancelling flux if they are approaching. Distances are measured in
is directed towards you, while the dark regions have magnetic fields arcsec, where 1 arcsec represents 726 km on the Sun (courtesy of
pointing away. The numbers indicate (1) the polar field, (2) a large- Sami Solanki).
scale unipolar field, (3) an active region, (4) a tiny bipolar region
(called an ephemeral region and a normal network), (5) a remnant
active region (6) and an enhanced network field. The two sunspot natural consequence of convection in the presence of a mag-
bands north and south of the equator can clearly be seen (from the netic field (Rempel, 2012, 2015).
HMI instrument on the Solar Dynamics Observatory, SDO).
3.2 The chromosphere

The IRIS satellite has revealed that the chromosphere is

very dynamic and has much fine structure. Spicules can
granules. These elements represent tiny magnetic flux tubes
be see at the limb of the Sun as cool jets of plasma that
poking through the photosphere and spreading out into the
are ejected from the network at the edges of supergranules.
chromosphere and the corona. The result is that the chromo-
They are heated to transition-region temperatures as they rise
spheric and coronal magnetic field is much more complex
(Skogsrud et al., 2015; Carlsson et al., 2015) and are often
than we once thought.
twisted (De Pontieu et al., 2014), but their contribution to the
Images of the photospheric magnetic field with the high-
solar wind is not yet known.
est spatial resolution have come from the wonderful Sunrise
The Bifrost code has been developed in Oslo to model the
balloon mission, which has been able to resolve the kilogauss
complex transition through the chromosphere to the corona
flux outside active regions for the first time and has a resolu-
(Martnez-Sykora et al., 2009; Hansteen et al., 2010, 2015;
tion of 100 km (Solanki et al., 2010; Solanki, 2017; Solanki
Carlsson et al., 2010; Gudiksen et al., 2011). It has provided
et al., 2017). An example is shown in Fig. 4, which is com-
new insight into the nature of the complex connection be-
pared with the previous highest resolution from the Hinode
tween the photosphere and the corona, which involves tran-
mission.
sitions between the forces from being pressure dominated to
Sunspots are locations of huge magnetic flux tubes with
those being magnetically dominated, the plasma from be-
magnetic fields generated at the base of the convection zone
ing neutral to being fully ionised, the radiation from be-
that then rise through the convection zone by so-called mag-
ing optically thick to being optically thin and the thermo-
netic buoyancy. The flux tubes give rise to sunspots when
dynamic state from being in local equilibrium to being in
they break through the solar surface, and they are dark be-
non-equilibrium.
cause they are cooler than the surrounding photosphere.
Granulation mixes up the surface layers with hotter layers
just below the surface and tends to make the surface hotter
and brighter than it would otherwise be. Inside a sunspot, 4 Magnetic reconnection in three dimensions
however, this process does not occur since the magnetic field
stops the granulation and makes the surface of the sunspot Before continuing to a discussion of coronal heating and so-
cooler and darker than the surrounding photosphere. lar flares, let us have a little detour to summarise the funda-
Observations of sunspots from Hinode and from the mental plasma process of magnetic reconnection, which is
Swedish Solar Telescope on La Palma have revealed fine likely to be involved in both topics (Priest and Forbes, 2000).
structure that would have delighted and intrigued Alfvn. The basic theory for the process in two dimensions is now
Also, recent computational models have produced amazingly well understood, but there are key differences when it comes
realistic images, which show that all the fine structure is a to three dimensions.

Ann. Geophys., 35, 805816, 2017 www.ann-geophys.net/35/805/2017/

E. Priest: Alfvn Medal lecture 809

Figure 6. (a) The structure in 2-D near an X-type null point due to
four sources on the base, showing how separatrix curves linking the
X-point separate the 2-D region into topologically separate regions.
Figure 5. (a) The magnetic structure near a 3-D null point where (b) The structure in 3-D due to four sources on the base, showing
the magnetic field vanishes. The spine is an isolated field line that how separatrix surfaces linking two null points (also on the base)
approaches or recedes from the null point. (b) The field lines for a separate the 3-D region into topologically separate volumes. The
null point with magnetic components (Bx , By , Bz ) = (x, y, 2z). two separatrices intersect in a special field line called a separator
that joins the one null point to the other.

First of all, the magnetic structure near a null point, at

which the magnetic field vanishes, is quite different in 2-D
and 3-D. In 2-D a null point forms either an X-type or an O-
type structure, but in 3-D the simplest field with a null at the
origin and satisfying B = 0 has components


Bx , By , Bz = (x, y, 2z), (6)

with field lines indicated in Fig. 5. Two families of field

lines link to the null point: an isolated spine field line, which
comes in along the positive and negative z axis towards the
null, and a surface of fan field lines, which comes out of the
null in the xy plane. The fan surface is a so-called sepa-
ratrix surface that separates the field lines that lie above the
xy plane from those that lie below it.
Secondly, the topology in 3-D is much more complex than Figure 7. The magnetic structure near a separator lying along the z
in 2-D. In 2-D, consider four sources in a line on the base axis and joining one null point to another.
(the photosphere) which are situated in the order +, , +,
(Fig. 6a). Above the base in the corona there is an X-type null
point. The field lines that thread the null point are called sep- The third difference between 2-D and 3-D lies in the nature
aratrix curves, since they separate the 2-D region into topo- of reconnection. In 2-D, reconnection can take place only at
logically distinct subregions in the sense that the field lines an X-point, and in Fig. 6a it transfers magnetic flux from
immediately to the left of the null all start out at the left- two subregions to the other two subregions. In 3-D, one type
most positive source and end at the leftmost negative source; of reconnection, called separator reconnection (Priest and
those to the right of the null link the other two sources. Fur- Titov, 1996; Longcope and Cowley, 1996; Galsgaard and
thermore, all the field lines below the null join the rightmost Nordlund, 1996; Longcope, 2001; Parnell and Galsgaard,
positive source to the leftmost negative source, while all the 2004; Parnell et al., 2010), occurs when a strong current
field lines above the null join the two outermost sources. builds up along a separator. It has the effect of transferring
In 3-D, by comparison, suppose two positive and two neg- magnetic flux from two regions to two other regions.
ative sources are placed on the plane base (photosphere), as However, in 3-D, reconnection can also take place in other
indicated in Fig. 6b. Two 3-D null points will be present on ways: at null points (Priest and Pontin, 2009), through tor-
the base, indicated by large dots, and the fan surfaces from sional spine reconnection, torsional fan reconnection or, most
each of those null points form domes that curve downwards usually, through spinefan reconnection (Pontin et al., 2005);
and separate the field lines below the dome from those above or quasi-separator reconnection when the current accumu-
it. When the two domes intersect, they do so in a special lates at a so-called quasi-separator (Priest and Dmoulin,
curve called a separator that joins the one null point to the 1995; Dmoulin et al., 1996; Titov et al., 2003; Aulanier
other (Fig. 7). et al., 2005). Consider the mapping of the footpoints of mag-
netic field lines from one part of the photosphere to another.

www.ann-geophys.net/35/805/2017/ Ann. Geophys., 35, 805816, 2017

810 E. Priest: Alfvn Medal lecture

If there is a 3-D null point in the overlying atmosphere, then z

there is a separatrix surface across which the mapping ex-
hibits a sudden jump or discontinuity. However, when there B2
is no null point or separatrix surface, there are sometimes so-
called quasi-separatrix surfaces, or QSLs for short, across B1
which the mapping is perfectly continuous but possesses a
steep gradient. Then large currents can once again accumu-
late along the quasi-separator and lead to magnetic reconnec-
tion.
An example of such a quasi-separator is shown in Fig. 8, y
for which the magnetic field just consists of an X-type null- A2
point field in xy planes and a uniform field (B0 ) in the z x
direction. The z component of the field therefore removes
the null points, but the separatrices that exist in the xy A1
and xz planes when B0 = 0 become quasi-separatrices when
B 6 = 0. Thus, for example, as a footpoint moves from B1 a Figure 8. The magnetic structure near a quasi-separator located
small distance to B2 , the other end of the field line flips very along the z axis.
rapidly over a large distance from A1 to A2 .
A final important new feature in three dimensions that we
shall use in Sect. 6 is the existence of a topological invari- field changes polarity. Occasionally, they erupt outwards in
ant known as magnetic helicity, which includes two types: association with and lying in the core of coronal mass ejec-
self-helicity and mutual helicity (Berger, 1984; Hornig and tions. When they erupt from active regions, they also produce
Schindler, 1996). The self-helicity is a measure of the twist- solar flares.
ing and kinking of a magnetic flux tube, whereas the mutual
helicity measures the linkage between different flux tubes. 5.1 Prominences
What happens during 3-D reconnection is that the total mag-
netic helicity (the sum of self-helicity and mutual helicity) is Recently, an impressive non-linear force-free model of the
conserved, but it can be converted from one kind to the other, coronal magnetic field has been set up (Mackay and van Bal-
for example, from mutual to self-helicity. legooijen, 2006; Mackay and Yeates, 2012; Yeates et al.,
2008; Yeates and Mackay, 2009). It evolves through a series
of equilibria in response to the observed surface motions at
5 The corona the Sun: differential rotation, meridional flow, flux diffusion
and emergence of new flux (Fig. 10a). Along certain por-
The corona has a temperature of about 1 million degrees K, tions of the global polarity inversion line, twisted magnetic
and in 1970 (apart from the occasional rocket flight) we could flux ropes appear; at the same locations, H maps reveal the
only glimpse it with one image per year from an eclipse ob- presence of prominences. The locations of the vast majority
servation. The corona was thought to be heated by sound of prominences that are either quiescent (i.e., outside active
waves generated by granulation, propagating upwards and regions) or intermediate (i.e., at the boundaries of active re-
then steepening to form shock waves. gions) can be predicted with the code (Fig. 10b). Further-
Now, in addition to eclipse photographs we have regu- more, the timing for the eruption of such prominences can be
lar images from artificial eclipses produced by space-borne predicted to within 1 day.
coronagraphs and can also view the corona direct regularly
in EUV and X-rays from space (Fig. 9). The corona has 5.2 Coronal heating
three different types of structure: small X-ray bright points
lying above opposite-polarity magnetic fragments in the pho- The coronal heating problem probably has a variety of differ-
tosphere, coronal loops of many different types and coronal ent solutions in different parts of the solar atmosphere. Thus,
holes through which the magnetic field is open to interplane- coronal holes and the high corona are likely to be heated by
tary space and along which the fast solar wind escapes. The magnetic waves propagating from below and dissipating by
slow solar wind comes from near the equatorial streamer belt. phase mixing or resonant absorption (Heyvaerts and Priest,
Key unanswered questions are related to the nature of 1983; Tomczyk and McIntosh, 2009; Nakariakov et al., 1999;
coronal heating (which is now known to be magnetic), the Goossens et al., 2011; De Moortel and Nakariakov, 2012),
acceleration of the solar wind and the nature of huge mag- with the dissipation mechanism being a collisionless one. X-
netic flux ropes called prominences, which lie in the corona ray bright points, on the other hand, are almost certainly pro-
but have chromospheric temperatures. Prominences lie above duced directly by magnetic reconnection, since they lie above
polarity inversion lines, at which the line-of-sight magnetic opposite-polarity magnetic sources that either move apart as

Ann. Geophys., 35, 805816, 2017 www.ann-geophys.net/35/805/2017/

E. Priest: Alfvn Medal lecture 811

Figure 9. The solar corona as viewed (a) in white light during an eclipse near solar maximum, courtesy of the High Altitude Observa-
tory (HAO), University Corporation for Atmospheric Research (UCAR) in Boulder, Colorado. UCAR is sponsored by the National Science
Foundation. (b) The solar corona in soft X-rays from the Yohkoh satellite (courtesy of Saku Tsuneta).

I
S

Figure 11. The coronal tectonics model, showing a schematic of

a coronal loop consisting of many sub-volumes, each linked to a
separate photospheric source and divided from one another by sep-
aratrix surfaces. The motions of the footpoints produce a myriad of
current sheets along the separatrices, which dissipate impulsively
Figure 10. (a) A non-linear force-free model of the solar corona and produce nano-flares.
from a global simulation after 108 days of evolution, showing
highly twisted flux ropes, weakly sheared arcades and near-potential
open fields. (b) A close-up showing a twisted flux rope lying above along the separatrices. These current sheets continually form
part of the polarity inversion line and (c) the associated solar promi- and dissipate through magnetic reconnection and produce a
nence observed by Big Bear Solar Observatory in H (courtesy of complex web of impulsive nano-flares that are highly likely
Duncan Mackay). to be heating the corona, especially in active regions and in
other types of coronal loop.

new flux emerges and reconnects with the overlying field or 5.3 The solar wind
come closer together and drive reconnection as they cancel
(Priest et al., 1994; Parnell and Priest, 1995). MHD turbu- In 1970, the focus was simply on Parker models of the solar
lence models may be used to describe the cascade of energy wind, with the flow velocity (v(r)) being spherically sym-
from large to small scales and its dissipation. metric and a function only of distance (r) from the Sun.
The coronal tectonics model (Priest et al., 2002) is a refine- Now the models are highly sophisticated, and observations
ment of Parkers braiding and his nano-flare model. Coronal from the Ulysses mission have revealed that at sunspot mini-
tectonics takes account of the fact that the magnetic field of mum there is a mixture of fast solar wind at 700 km s1 from
the corona comes through the solar surface from many small coronal holes and much more sporadic slow solar wind from
intense photospheric sources rather than through regions of streamer belts. On the other hand, at sunspot maximum the
weak uniform magnetic field. Thus, the flux from the sources solar wind is much more irregular (Fig. 12).
spreads out in a highly complex way into the overlying at- The fast solar wind is likely to originate in coronal funnels,
mosphere, and the fluxes from neighbouring sources are sep- as shown in Hinode observations of outflows (Sakao et al.,
arated from one another by a complex web of separatrix sur- 2007). Models of the magnetic topology of the corona have
faces (Fig. 11). Thus, the response to the complex motions of revealed a series of funnels bounded by open separatrices and
these sources is the formation of a myriad of current sheets often spreading out from small locations at the solar surface

www.ann-geophys.net/35/805/2017/ Ann. Geophys., 35, 805816, 2017

812 E. Priest: Alfvn Medal lecture

Figure 12. The speed of the solar wind as a function of latitude in Figure 13. (a) An erupting prominence in He II 304 with the size
polar coordinates from the Ulysses spacecraft during (a) solar min- of the Earth indicated (from the AIA instrument on SDO). (b) A
imum and (b) solar maximum. Superimposed are coronal images coronal mass ejection observed by the LASCO instrument on-board
from the Mauna Loa K coronameter, plus the EIT and LASCO in- SOHO.
struments on the Solar Dynamics Observatory (SDO) (from Meyer-
Vernet, 2007).

(Platten et al., 2014; Edwards et al., 2016). So the fast wind

is in my view likely to be heated by reconnection low down
in funnels and then by a turbulent cascade of MHD waves
higher up (Cranmer, 2012). On the other hand, the mecha-
nism for accelerating the slow solar wind is much more un-
certain. One possibility is that it is heated by the equivalent
of the tectonics model, but operating in a web of separatrices
near streamer boundaries (the so-called S-web) (Linker et al.,
2011; Titov et al., 2011).

Figure 14. (a) Schematics of a solar flare, showing (a) the 3-D
6 Solar flares and coronal mass ejections structure and (b) a vertical cross section early in the main phase
and (c) later on.
In 1970, a solar flare was defined as a brightening in the
chromosphere, and coronal mass ejections (CMEs) had not
been born. Now we know that the core of a flare is located in The cause of the eruption may be a lack of equilibrium,
the corona and involves the eruption of a coronal magnetic called non-equilibrium, or it may be a torus or kink instabil-
flux tube, which drives reconnection below it. The magnetic ity. It appears that in some flares, separator reconnection is
flux tube is revealed as an erupting prominence and the sur- at work, while others may be focussed on quasi-separators
rounding coronal structure becomes a CME (Fig. 13). or on coronal null points. Some of the twist that is observed
A schematic of the main phase of the overall process in the erupting flux rope and further out in magnetic clouds
is shown in Fig. 14a; a vertical cross section through it may come from an initial flux tube located below the coro-
(Fig. 14b) indicates the nature of the reconnection that is nal arcade before it erupts. However, we have just suggested
driven below the erupting prominence. The heat and fast par- a new process that may create most of the twist (Priest and
ticles that are produced during the reconnection propagate Longcope, 2017).
down to the feet of the reconnected field lines and drive chro- During the rise phase of a flare, there are chromospheric
mospheric upflows, which fill a coronal loop with hot dense brightenings at small points on either side of the polarity in-
plasma through a process called chromospheric evaporation. version line (PIL), which then spread along the PIL to create
At the base of the resulting hot coronal arcade, one sees a the ribbons. We suggest, therefore, that the initial sheared
pair of bright chromospheric ribbons. As the reconnection arcade (overlying an initial small flux rope Z+ Z ) consists
proceeds, the location of the reconnection rises in the atmo- of a series of flux tubes A+ A , B+ B , C+ C and D+ D
sphere, so new higher coronal loops are heated (Fig. 14c) (Fig. 15a). The flare in our model spreads along the PIL by
while the ribbons separate in time. Meanwhile, the previous so-called zipper reconnection, in which A+ A first recon-
coronal loops cool through radiation and heat conduction. nects with B+ B (Fig. 15b), then with C+ C (Fig. 15c) and
The reconnection adds both magnetic flux and magnetic he- finally with D+ D (Fig. 15d).
licity to the erupting flux rope. At each reconnection, the total magnetic helicity is con-
served, but the mutual magnetic helicity of the sheared (in-

Ann. Geophys., 35, 805816, 2017 www.ann-geophys.net/35/805/2017/

E. Priest: Alfvn Medal lecture 813

Figure 15. Zipper reconnection in a coronal arcade, seen from above, during the rise phase of a flare or coronal mass ejection. (a) Suppose
the initial arcade consists of four parallel flux tubes connecting sources A+ A , B+ B , C+ C and D+ D . These overlie
a pre-existing flux rope Z+ Z . The zipper reconnection sequence consists of three individual reconnection events (bd). The overlying
twisted flux tube is designated R, and the part of the reconnected arcade tube that is generated in reconnection event n is designated with that
number, i.e., 1, 2 or 3.

ner) arcade is converted into self-helicity of the new flux

rope. Thus, after the first reconnection, the reconnected field
line A+ B is twisted about the initial tube Z+ Z . After sev-
eral reconnections, the end effect of the zipper process is to
create a core of high twist A+ D around the initial tube.
During the main phase of the flare, the flux rope contin-
ues to grow by quasi-2-D reconnection, with mutual helicity
of the outer coronal arcade being converted to self-helicity
of the outer part of the flux rope (Fig. 16). The net effect
is that the final erupting flux rope consists of three parts: a
central inner core that was present initially, a highly twisted
outer core that was created by zipper reconnection during the
rise phase and an outer less-twisted sheath created during the
main phase.

Figure 16. Quasi-2-D reconnection during the main phase of a flare

7 Conclusions or coronal mass ejection. (a) The initial sheared state. (b) The situ-
ation after the zipper phase of the flare when reconnection has pro-
Solar physics is at an extremely exciting phase, since there ceeded parallel to the polarity inversion line to produce the zipper
are many intriguing, fundamental and unsolved problems. flux rope (A+ D ) lying under the next set of field lines joining
Huge progress has been made over the past 10 years which footpoints E+ E , F+ F , G+ G , H+ H . (c) The first part of the
would have delighted Alfvn, but definitive detailed answers main phase after reconnection has progressed sideways to reconnect
have not yet been provided about the following: the next sheath of field lines and create a spiral sheath that wraps
around the zipper flux rope and enhances its flux and magnetic he-
how the solar magnetic field is generated; licity.
how the dynamic fine-scale structure of the solar atmo-
sphere is created;
Future breakthroughs will come from a combination of new
how the different parts of the corona are heated; observations, especially from the European Solar Orbiter
mission (2019) and the DKIST 4 m telescope on Hawaii,
how the different types of solar wind are accelerated; but also from clever new ideas and models in the spirit of
Hannes Alfvn. So, in the meantime let us relish the chal-
how eruptive solar flares and coronal mass ejections are
lenge and enjoy the beauty of the Sun.
initiated;
and how reconnection converts magnetic energy to ki-
netic energy, heat and fast-particle energy in flares.

www.ann-geophys.net/35/805/2017/ Ann. Geophys., 35, 805816, 2017

814 E. Priest: Alfvn Medal lecture

Data availability. No data sets were used in this article. P., De Luca, E. E., Golub, L., McKillop, S., Reeves, K., Saar,
S., Testa, P., Tian, H., Kankelborg, C., Jaeggli, S., Kleint, L.,
and Martinez-Sykora, J.: On the prevalence of small-scale twist
Competing interests. The author declares that he has no conflict of in the solar chromosphere and transition region, Science, 346,
interest. 1255732, https://doi.org/10.1126/science.1255732, 2014.
Dmoulin, P., Henoux, J., Priest, E. R., and Mandrini, C. H.: Quasi-
separatrix layers in solar flares. I. Method, Astron. Astrophys.,
Acknowledgements. I am delighted to acknowledge the fun that 308, 643655, 1996.
many collaborators, colleagues, students and postdocs have given Dungey, J.: Interplanetary magnetic field and
me as we have shared ideas over the years. Without them and the the auroral zones, Phys. Rev. Lett., 6, 4748,
unceasing support of my wife, Clare, and my family I would not https://doi.org/10.1103/PhysRevLett.6.47, 1961.
have been able to embark on the many research journeys. Duvall, T. L. and Harvey, J. W.: Rotational frequency
The topical editor, Manuela Temmer, thanks two anonymous splitting of solar oscillations, Nature, 310, 1922,
referees for help in evaluating this paper. https://doi.org/10.1038/310019a0, 1984.
Duvall, T. L., Dziembowski, W. A., Goode, P. R., Gough, D. O.,
Harvey, J. W., and Leibacher, J. W.: Internal rotation of the Sun,
Nature, 310, 2225, https://doi.org/10.1038/310022a0, 1984.
References Edwards, S. J., Parnell, C. E., Harra, L. K., Culhane, J. L., and
Brooks, D. H.: A Comparison of Global Magnetic Field Skele-
Archontis, V. and Hansteen, V.: Clusters of Small Eruptive Flares tons and Active-Region Upflows, Solar Phys., 291, 117142,
Produced by Magnetic Reconnection in the Sun, Astrophys. https://doi.org/10.1007/s11207-015-0807-8, 2016.
J. Lett., 788, L2, https://doi.org/10.1088/2041-8205/788/1/L2, Galsgaard, K. and Nordlund, .: The heating and activity of
2014. the solar corona: I Boundary shearing of an initially homo-
Archontis, V., Hood, A. W., and Tsinganos, K.: Recurrent Explosive geneous magnetic field, J. Geophys. Res., 101, 1344513460,
Eruptions and the Sigmoid-to-arcade Transformation in the https://doi.org/10.1029/96JA00428, 1996.
Sun Driven by Dynamical Magnetic Flux Emergence, Astrophys. Goossens, M., Erdlyi, R., and Ruderman, M. S.: Resonant MHD
J. Lett., 786, L21, https://doi.org/10.1088/2041-8205/786/2/L21, Waves in the Solar Atmosphere, Space Sci. Rev., 158, 289338,
2014. https://doi.org/10.1007/s11214-010-9702-7, 2011.
Aulanier, G., Pariat, E., and Dmoulin, P.: Current-sheet forma- Gudiksen, B. V., Carlsson, M., Hansteen, V. H., Hayek, W.,
tion in quasi-separatrix layers and hyperbolic flux tubes, As- Leenaarts, J., and Martnez-Sykora, J.: The stellar atmo-
tron. Astrophys., 444, 961976, https://doi.org/10.1051/0004- sphere simulation code Bifrost. Code description and validation,
6361:20053600, 2005. Astron. Astrophys., 531, A154, https://doi.org/10.1051/0004-
Berger, M. A.: Rigorous new limits on magnetic helicity dissipation 6361/201116520, 2011.
in the solar corona, Geophys. Astrophys. Fluid Dynam., 30, 79 Hansteen, V. H., Hara, H., De Pontieu, B., and Carlsson, M.: On
104, https://doi.org/10.1080/03091928408210078, 1984. Redshifts and Blueshifts in the Transition Region and Corona,
Birkeland, K.: The Norwegian Aurora Polaris Expedition, 1902 Astrophys. J., 718, 10701078, https://doi.org/10.1088/0004-
1903, in: vol. I, Sect. 1, H. Aschehoug, Oslo, 1908. 637X/718/2/1070, 2010.
Carlsson, M., Hansteen, V. H., and Gudiksen, B. V.: Chromospheric Hansteen, V. H., Guerreiro, N., De Pontieu, B., and Carlsson, M.:
heating and structure as determined from high resolution 3D sim- Numerical Simulations of Coronal Heating through Footpoint
ulations, Mem. della Soc. Astron. Ital., 81, 582587, 2010. Braiding, Astrophys. J., 811, 106, https://doi.org/10.1088/0004-
Carlsson, M., Leenaarts, J., and De Pontieu, B.: What Do 637X/811/2/106, 2015.
IRIS Observations of Mg II k Tell Us about the So- Hansteen, V. H., Archontis, V., Pereira, T. M. D., Carlsson, M.,
lar Plage Chromosphere?, Astrophys. J. Lett., 809, L30, Rouppe van der Voort, L., and Leenaarts, J.: Bombs and Flares
https://doi.org/10.1088/2041-8205/809/2/L30, 2015. at the Surface and Lower Atmosphere of the Sun, Astrophys. J.,
Chapman, S. and Ferraro, V. C. A.: A New The- 839, 22, https://doi.org/10.3847/1538-4357/aa6844, 2017.
ory of Magnetic Storms, Nature, 126, 129130, Heyvaerts, J. and Priest, E. R.: Coronal heating by phase-mixed
https://doi.org/10.1038/126129a0, 1930. shear Alfvn waves, Astron. Astrophys., 117, 220234, 1983.
Christensen-Dalsgaard, J., Duvall, T. L., Gough, D. O., Harvey , Hornig, G. and Schindler, K.: Magnetic topology and the prob-
J. W., and Rhodes, E. J.: Speed of sound in the solar interior, lem of its invariant definition, Phys. Plasmas, 3, 781791,
Nature, 315, 378382, https://doi.org/10.1038/315378a0, 1985. https://doi.org/10.1063/1.871778, 1996.
Cranmer, S. R.: Self-Consistent Models of the Solar Wind, Space Linker, J. A., Lionello, R., Mikic, Z., Titov, V. S., and
Sci. Rev., 172, 145156, https://doi.org/10.1007/s11214-010- Antiochos, S. K.: The Evolution of Open Magnetic Flux
9674-7, 2012. Driven by Photospheric Dynamics, Astrophys. J., 731, 110,
De Moortel, I. and Nakariakov, V. M.: Magnetohydrodynamic https://doi.org/10.1088/0004-637X/731/2/110, 2011.
waves and coronal seismology: an overview of recent re- Longcope, D. W.: Separator current sheets: Generic
sults, Philos. T. Roy. Soc. Lond. A, 370, 31933216, features in minimum-energy magnetic fields sub-
https://doi.org/10.1098/rsta.2011.0640, 2012. ject to flux constraints, Phys. Plasmas, 8, 52775289,
De Pontieu, B., Rouppe van der Voort, L., McIntosh, S. W., Pereira, https://doi.org/10.1063/1.1418431, 2001.
T. M. D., Carlsson, M., Hansteen, V., Skogsrud, H., Lemen, J.,
Title, A., Boerner, P., Hurlburt, N., Tarbell, T. D., Wuelser, J.

Ann. Geophys., 35, 805816, 2017 www.ann-geophys.net/35/805/2017/

E. Priest: Alfvn Medal lecture 815

Longcope, D. W. and Cowley, S. C.: Current sheet formation along Priest, E. R. and Titov, V. S.: Magnetic reconnection at three-
three-dimensional magnetic separators, Phys. Plasmas, 3, 2885 dimensional null points, Philos. T. Roy. Soc. Lond., 355, 2951
2897, https://doi.org/10.1063/1.871627, 1996. 2992, https://doi.org/10.1098/rsta.1996.0136, 1996.
Mackay, D. H. and van Ballegooijen, A. A.: Models of the Priest, E. R., Parnell, C. E., and Martin, S. F.: A converg-
Large-Scale Corona. I. Formation, Evolution, and Liftoff ing flux model of an X-ray bright point and an associ-
of Magnetic Flux Ropes, Astrophys. J., 641, 577589, ated canceling magnetic feature, Astrophys. J., 427, 459474,
https://doi.org/10.1086/500425, 2006. https://doi.org/10.1086/174157, 1994.
Mackay, D. H. and Yeates, A.: The Suns Global Photospheric and Priest, E. R., Heyvaerts, J., and Title, A.: A Flux Tube Tectonics
Coronal Magnetic Fields: Observations and Models, Liv. Rev. model for solar coronal heating driven by the magnetic carpet,
Sol. Phys., 9, 6, https://doi.org/10.12942/lrsp-2012-6, 2012. Astrophys. J., 576, 533551, https://doi.org/10.1086/341539,
Martnez-Sykora, J., Hansteen, V., DePontieu, B., and Carlsson, 2002.
M.: Spicule-Like Structures Observed in Three-Dimensional Re- Rempel, M.: Numerical Sunspot Models: robustness of Photo-
alistic Magnetohydrodynamic Simulations, Astrophys. J., 701, spheric Velocity and Magnetic Field Structure, Astrophys. J.,
15691581, https://doi.org/10.1088/0004-637X/701/2/1569, 750, 62, https://doi.org/10.1088/0004-637X/750/1/62, 2012.
2009. Rempel, M.: Numerical Simulations of Sunspot Decay: On the
Martnez-Sykora, J., Moreno-Insertis, F., and Cheung, M. C. M.: Penumbra-Evershed Flow-Moat Flow Connection, Astrophys. J.,
Multi-parametric Study of Rising 3D Buoyant Flux Tubes in 814, 125, https://doi.org/10.1088/0004-637X/814/2/125, 2015.
an Adiabatic Stratification Using AMR, Astrophys. J., 814, 2, Sakao, T., Kano, R., Narukage, N., Kotoku, J., Bando, T., DeLuca,
https://doi.org/10.1088/0004-637X/814/1/2, 2015. E. E., Lundquist, L. L., Tsuneta, S., Harra, L. K., Katsukawa, Y.,
Meyer-Vernet, N.: Basics of the Solar Wind, Cambridge University Kubo, M., Hara, H., Matsuzaki, K., Shimojo, M., Bookbinder, J.
Press, Cambridge, UK, 2007. A., Golub, L., Korreck, K. E., Su, Y., Shibasaki, K., Shimizu, T.,
Nakariakov, V. M., Ofman, L., DeLuca, E. E., Roberts, B., and and Nakatani, I.: Continuous Plasma Outflows from the Edge of a
Davila, J. M.: TRACE observation of damped coronal loop oscil- Solar Active Region as a Possible Source of Solar Wind, Science,
lations: Implications for coronal heating, Science, 285, 862864, 318, 15851588, https://doi.org/10.1126/science.1147292, 2007.
https://doi.org/10.1126/science.285.5429.862, 1999. Schou, J., Antia, H. M., Basu, S., Bogart, R. S., Bush, R. I., Chitre,
Parnell, C. E. and Galsgaard, K.: Elementary Heating Events S. M., Christensen-Dalsgaard, J., di Mauro, M. P., Dziembowski,
Interaction between Two Flux Sources. II Rates of W. A., Eff-Darwich, A., Gough, D. O., Haber, D. A., Hoeksema,
Flux Reconnection, Astron. Astrophys., 428, 595612, J. T., Howe, R., Korzennik, S. G., Kosovichev, A. G., Larsen, R.
https://doi.org/10.1051/0004-6361:20034350, 2004. M., Pijpers, F. P., Scherrer, P. H., Sekii, T., Tarbell, T. D., Title,
Parnell, C. E. and Priest, E. R.: A converging flux model for A. M., Thompson, M. J., and Toomre, J.: Helioseismic Studies of
the formation of an X-ray bright point above a supergran- Differential Rotation in the Solar Envelope by the Solar Oscilla-
ule cell, Geophys. Astrophys. Fluid Dynam., 80, 255276, tions Investigation Using the Michelson Doppler Imager, Astro-
https://doi.org/10.1080/03091929508228958, 1995. phys. J., 505, 390417, https://doi.org/10.1086/306146, 1998.
Parnell, C. E., Haynes, A. L., and Galsgaard, K.: Structure of mag- Skogsrud, H., Rouppe van der Voort, L., De Pontieu, B., and
netic separators and separator reconnection, J. Geophys. Res., Pereira, T. M. D.: On the Temporal Evolution of Spicules Ob-
115, 2102, https://doi.org/10.1029/2009JA014557, 2010. served with IRIS, SDO, and Hinode, Astrophys. J., 806, 170,
Platten, S. J., Parnell, C. E., Haynes, A. L., Priest, E. R., and https://doi.org/10.1088/0004-637X/806/2/170, 2015.
Mackay, D. H.: The solar cycle variation of topological struc- Solanki, S. K.: Overview of the Special Issue on the First Science
tures in the global solar corona, Astron. Astrophys., 565, A44, Results from the Second Flight of Sunrise, Ap. J. Suppl., 229, 1,
https://doi.org/10.1051/0004-6361/201323048, 2014. https://doi.org/10.3847/1538-4365/aa6742, 2017.
Pontin, D. I., Galsgaard, K., Hornig, G., and Priest, E. R.: A fully Solanki, S. K., Barthol, P., Danilovic, S., Feller, A., Gandorfer, A.,
magnetohydrodynamic simulation of 3D non-null reconnection, Hirzberger, J., Riethmller, T. L., Schssler, M., Bonet, J. A.,
Phys. Plasmas, 12, 052307, https://doi.org/10.1063/1.1891005, Martnez Pillet, V., del Toro Iniesta, J. C., Domingo, V., Palacios,
2005. J., Knlker, M., Bello Gonzlez, N., Berkefeld, T., Franz, M.,
Priest, E. R. and Dmoulin, P.: Three-dimensional mag- Schmidt, W., and Title, A. M.: SUNRISE: Instrument, Mission,
netic reconnection without null points. 1 Basic theory of Data, and First Results, Astrophys. J. Lett., 723, L127L133,
magnetic flipping, J. Geophys. Res., 100, 2344323463, https://doi.org/10.1088/2041-8205/723/2/L127, 2010.
https://doi.org/10.1029/95JA02740, 1995. Solanki, S. K., Riethmller, T. L., Barthol, P., Danilovic, S.,
Priest, E. R. and Forbes, T. G.: Magnetic Reconnection: MHD The- Deutsch, W., Doerr, H.-P., Feller, A., Gandorfer, A., Germerott,
ory and Applications, Cambridge University Press, Cambridge, D., Gizon, L., Grauf, B., Heerlein, K., Hirzberger, J., Kolleck,
UK, 2000. M., Lagg, A., Meller, R., Tomasch, G., van Noort, M., Blanco
Priest, E. R. and Longcope, D. W.: Flux-Rope Twist in Eruptive Rodrguez, J., Gasent Blesa, J. L., Balaguer Jimnez, M., Del
Flares and CMEs: Due to Zipper and Main-Phase Reconnection, Toro Iniesta, J. C., Lpez Jimnez, A. C., Orozco Suarez, D.,
Solar Phys., 292, 25, https://doi.org/10.1007/s11207-016-1049- Berkefeld, T., Halbgewachs, C., Schmidt, W., lvarez-Herrero,
0, 2017. A., Sabau-Graziati, L., Prez Grande, I., Martnez Pillet, V.,
Priest, E. R. and Pontin, D. I.: Three-dimensional Card, G., Centeno, R., Knlker, M., and Lecinski, A.: The
null point regimes, Phys. Plasmas, 16, 122101, Second Flight of the Sunrise Balloon-borne Solar Observatory:
https://doi.org/10.1063/1.3257901, 2009. Overview of Instrument Updates, the Flight, the Data, and First

www.ann-geophys.net/35/805/2017/ Ann. Geophys., 35, 805816, 2017

816 E. Priest: Alfvn Medal lecture

Results, Ap. J. Supplement, 229, 2, https://doi.org/10.3847/1538- Tomczyk, S. and McIntosh, S. W.: Time-Distance Seismology of
4365/229/1/2, 2017 the Solar Corona with CoMP, Astrophys. J., 697, 13841391,
Southwood, D.: From the Carrington Storm to the Dungey Mag- https://doi.org/10.1088/0004-637X/697/2/1384, 2009.
netosphere, in: Magnetospheric Plasma Physics: The Impact of Yeates, A. R. and Mackay, D. H.: Initiation of Coronal Mass Ejec-
Jim Dungeys Research, vol. 41 of Astrophysics and Space Sci- tions in a Global Evolution Model, Astrophys. J., 699, 1024
ence Proceedings, edited by: Southwood, D., Cowley, S. W. H., 1037, https://doi.org/10.1088/0004-637X/699/2/1024, 2009.
and Mitton, S., p. 253, https://doi.org/10.1007/978-3-319-18359- Yeates, A. R., Mackay, D. H., and van Ballegooijen, A.
6_12, Springer International Publishing Switzerland, 2015. A.: Evolution and Distribution of Current Helicity in Full-
Titov, V. S., Galsgaard, K., and Neukirch, T.: Magnetic pinching of Sun Simulations, Astrophys. J. Lett., 680, L165L168,
hyperbolic flux tubes. I. Basic estimations, Astrophys. J., 582, https://doi.org/10.1086/590057, 2008.
11721189, https://doi.org/10.1086/344799, 2003.
Titov, V. S., Mikic, Z., Linker, J. A., Lionello, R., and Antiochos, S.
K.: Magnetic Topology of Coronal Hole Linkages, Astrophys. J.,
731, 111, https://doi.org/10.1088/0004-637X/731/2/111, 2011.

Ann. Geophys., 35, 805816, 2017 www.ann-geophys.net/35/805/2017/

Copyright of Annales Geophysicae (09927689) is the property of Copernicus Gesellschaft
mbH and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.