Low Coronal Signatures

1
MULLARD SPACE SCIENCE LABORATORY DEPT SPACE AND
CLIMATE PHYSICS
Low Coronal Signatures of Coronal Mass
Ejections Coronal Waves and Dimmings
Gemma Attrill
2
Coronal Mass Ejections (CMEs)

• Mass 1016 g material from the low corona.
• Magnetic flux expelled 1023 Mx
• Velocities lt200 to gt3000 km s-1.
• (Webb et al., 1994 Gosling, 1990 Williams et
  al., 2005)
• Typical energy 1025 J

• Space weather

• First observed from space in 1973 with
  coronagraphs on the Orbiting Solar Observatory
  OSO-7 and Skylab. (MacQueen et al., 1974, Hildner
  et al., 1975, Gosling et al., 1976)

• Coronagraph occulting disc (or eclipse) blocks
  the near surface manifestations of the CME.
• Need low corona observations to study the source
  regions.

3
Coronal Dimming
What are coronal dimmings?

• Dimming is seen as a decrease in intensity in
  both EUV and X-ray images (e.g. Thompson et al.,
  1998 Sterling Hudson, 1997).

• First space-based observation by Skylab mission
  (1973-74) Transient Coronal Holes (Rust, 1983)

4
Coronal Waves

• Discovered by SOHO/EIT EIT waves (Dere et
  al., 1997 Thompson et al., 1998).

Coronal waves have a strong association with CMEs
(Biesecker et al. 2002) From a sample of 173
coronal wave events in every case a coronal
mass ejection occurred in association with the
EIT wave.
POSSIBLE CAUSES Many and varied!
5
Coronal Dimming

• Chapter 5 Early-stage evolution of coronal
  dimmings
• Attrill et al., (2006), Sol. Phys., 238, 117.
• Chapter 6 Late-stage recovery of coronal
  dimmings
• Attrill et al., (2008), Sol. Phys., in press.

6
Early-Stage Evolution of Coronal Dimmings (Ch 5)
Motivation Study the magnetic nature of
dimmings, since we want to understand the
magnetic nature of CMEs.

• Measured magnetic flux through the dimmings,
  improving on earlier studies by correcting for
  MDIs underestimation of the magnetic flux.

• Global context examined mysterious brightening
  (Aurora Solaris Hudson et al., 1996) along,
  and shrinking of, north polar coronal hole
  boundary.

7
Early-Stage Evolution of Coronal Dimmings (Ch 5)

• Derived a new interpretation of the event from
  the solar observations.

Summary (i) Showed for the first time that
study of the evolution of coronal dimmings can be
used to probe the global magnetic connectivity of
the CME post-eruption. (ii) Azimuthal component
important in calculations of MC flux.
8
Late-Stage Recovery of Coronal Dimmings (Ch 6)

• Motivation Why and how do coronal dimmings
  disappear whilst the magnetic connectivity of the
  CME ejecta to the Sun is maintained?

Question posed by Kahler Hudson (2001), along
with our observation that the average intensity
of the 12th May 1997 dimmings recover within 48
hours, whilst suprathermal uni-directional
electron heat fluxes are observed at 1 AU in the
related ICME more than 70 hours after the
eruption.

• Studied 3 events 12th May 1997, 13th May 2005
  and 6th July 2006.

• Found that recovery occurred by internal
  brightening as well as by shrinking.
• Combined in-situ and solar analysis revealed
  that recovery does not necessarily
  disconnection.

9
Late-Stage Recovery of Coronal Dimmings (Ch 6)
Fisk and Zurbuchen (2006)

• Showed for the first time that EUV dimmings have
  a 3D structure.

• Consider that dimmings recover by a dispersal
  of open magnetic flux.
• Showed that diffusion due to convective
  motions alone inadequate to explain recovery.

• However, model of interchange reconnection
  between surrounding quiet Sun closed loops and
  emerging flux would recover the dimming, whilst
  maintaining the magnetic connectivity of the
  ejecta to the Sun.
• Model of Fisk and Schwadron (2001), ? 1.6 x
  105 km2 s-1. (c.f. dArea/dTime 2.6 x 105 km2
  s-1).

Summary First work with EUV data on recovery of
dimmings. Showed that recovery can be due to
interchange reconnections facilitating dispersal
(as opposed to disconnection) of open flux.
10
Coronal Waves

• Chapter 3 Understanding Coronal Waves
• Literature review of both observations and
  theory.
• Chapter 4 Coronal Wave Magnetic Footprint of
  a CME?
• Attrill et al., (2007), ApJ, 656, L101.
• Attrill et al., (2007), Astron. Nachr., 328, 760.

11
Understanding coronal waves (Ch 3)
Motivation Clearly strongly associated with
CMEs, but no consensus on what they are

• Fast-mode MHD wave/shock (flare-induced
  blast-wave CME piston-driven wave)1
• Compressive -gt brightness enhancement.
• Can travel ? to the magnetic field.

• Plasma compression4
• Expansion of magnetic field during CME lift-off
  compresses plasma between stable flux domains.
• Stationary brightenings.
• Coronal dimming.

• Slow-mode MHD wave/shock2
• Compressive -gt brightness enhancement.
• Velocities lt vA

• Electric Currents5
• - Generated in large-scale QSL about twisted flux
  tube, as CME pushes overlying magnetic field.
• Line-of-sight integration produces on-disk
  bright front.

• Solitary wave3
• Can travel at a wide range of velocities.
• The most well-defined (density enhanced) coronal
  waves have a tendency to travel faster.

1 Dere et al., 1997 Thompson et al., 1998, 1999,
2000 Cliver et al., 1999 Wills-Davey
Thompson, 1999 Mann et al., 1999 Wang, 2000
Klassen et al., 2000 Gopalswamy Thompson,
2000 Wu et al., 2001Khan Aurass, 2002 Ofman
Thompson, 2002 Vrsnak et al., 2002 Warmuth et
al., 2004 Gilbert Holzer, 2004 Ballai et al.,
2005 Warmuth et al., 2005 Vr?snak et al., 2005
Cliver et al., 2005 Veronig et al., 2006,
Warmuth, 2007 Pomoell et al., 2008 Ballai et
al., 2008. 2 Krasnoselskikh Podladchikova,
2007. 3 Wills-Davey et al., 2007. 4 Delannee
Aulanier, 1999 Delannee, 2000 Chen et al.,
2002, 2005. 5 Delannee et al., 2008.
12
Understanding Coronal Waves (Ch 3)
Further complicated by two morphologically
different types of coronal wave (Thompson et
al., 2000 Biesecker et al., 2002 Vrsnak, 2005)

• Why so many interpretations? - Many different
  observed characteristics.

• Analysed evidence in the literature to date for
  possible relationship to observations from other
  spectral ranges radio data, H-alpha, X-ray, He I.

Summary Good evidence for relationship between
S-waves and type II radio bursts, H-alpha Moreton
waves, SXT waves and He I waves. -gt
S-waves understood to be fast-mode MHD waves.
But S-waves only constitute 7 coronal waves
(Biesecker et al., 2002).
Diffuse bright fronts remain an enigma.
13
Coronal wave Magnetic Footprint of a CME? (Ch
4)
Motivation Discovery that a diffuse coronal
wave rotates as it expands. (Podladchikova
Berghmans, 2005). Controversial result!

• Independently confirmed this result and went on
  to show that sense of rotation depends on the
  helicity of the CME source region. -gt Diffuse
  coronal wave cannot be a flare-induced blast-wave.

• Rotation of coronal wave in same sense as
  erupting filament rotation.

• 180o separation of peaks -gt consistent with
  interpretation in terms of a flux rope structure.

• Shift of bright front toward centre of disk
  during expansion. Consistent with projection
  effect if bright front is located in low corona.

14
Coronal wave Magnetic Footprint of a CME? (Ch
4)

• New model proposed for generation of diffuse
  bright fronts and two types of dimmings.
• Bright front generated by series of successive
  weak flare-like reconnection events. Dimmings
  due to plasma expanding into a larger volume.

Interaction between expanding CME and surrounding
magnetic polarities important.

• Summary
• Coronal wave is magnetic footprint of CME
  bubble in the low corona. It is driven by the
  erupting magnetic configuration. The diffuse
  bright front in this model is not a true wave
  (either flare-initiated or CME-driven).

• Bright front exists if the interface between
  CME and surroundings is favourable for magnetic
  reconnection. Vanishes if unfavourable.
• Formation of bright front preceeds dimmings
  (reconnection allows opening).
• Core dimmings remain rooted near source region,
  secondary dimmings expand following bright front.
• Every CME that expands laterally in the low
  corona should generate a coronal wave in the
  surrounding quiet Sun.

15
Coronal wave Magnetic Footprint of a CME? (Ch
4)
Tackling some big questions and challenging
established concepts

• Every EIT wave is associated with a CME, why
  doesnt every CME have an EIT wave?

- Studied all (81) large-scale limb CMEs January
1997 - June 1998.

• All (55) large-scale CMEs with a front-side
  source region have associated diffuse coronal
  waves.

Only a few coronal waves are semi-isotropic.
- Re-analysed EIT wave catalogue events
(Thompson Myers, 2008).

• All coronal waves identified with gt75
  confidence level are semi-isotropic.

The source region of a coronal wave is commonly
understood to be an active region transient.
- The eruption of a quiescent filament can drive
a diffuse coronal wave.

• The required component for driving a diffuse
  coronal wave is an erupting flux rope.

Coronagraph observations suggest that the
horizontal scale of the opened field can be many
times greater than that of the reconnection
arcade, and this may be difficult to reconcile
with the geometry of the existing models.
(Klimchuk, 2001).

• Our model allows us to understand how CMEs can
  become much larger than the initial erupting
  configuration in the low corona.

16
Summary

• Dimmings
• Dimming evolution -gt probe the large-scale
  magnetic connectivity and evolution of a CME,
  post-eruption.
• Dimmings may recover by a dispersal, rather than
  a disconnection, of open magnetic flux.

• Coronal waves
• Significant new observational constraints
  diffuse bright front has a magnetic nature.
  Existing models (waves, compression, current
  shell) fail to explain this.
• All confidently identified diffuse coronal
  waves are semi-isotropic.
• Every large-scale CME generates a diffuse
  coronal wave.
• Erupting magnetic flux rope required to drive
  a diffuse coronal wave.
• Developed a new model. When confronted with
  the many various observations to date, it
  survives remarkably well.
• Enhances our understanding of how CMEs can
  become large-scale even in the low corona.

Understand coronal waves dimmings as low
coronal signatures of CMEs.
17
Future work

• Dynamic numerical simulation and testing of our
  model.
• Observational analysis of SDO/AIA high cadence,
  high spatial resolution EUV data in many
  different wavelengths simultaneously.