Coronal Mass Ejections

1
Coronal Mass Ejections

• Nat Gopalswamy
• NASA Goddard Space Flight Center, Greenbelt,
  Maryland
• Presented at the International Symposium on Solar
  Activity, Weihai, 2002
• 1. Overview
• 2. Observational Properties
• 3. Associated Phenomena
• 4. CME Models
• 5. Current developments Future Perspectives

2
Coronal Mass Ejections 1 An Overview

• What are CMEs?
• Topology of solar magnetic fields
• Why study CMEs?
• Sun-Earth Connections Historical milestones
• Driver Gas, Shocks, SEPs
• Flares and Prominences
• IPS and in situ Observations

3
Defining a CME SOHO/LASCO/C3 Images
FRONTAL
CAVITY
CORE
Corona before CME Corona
with CME
4
Another example The 1997/02/07 CME (Gopalswamy
et al, 1998, GRL 25, 2485)

• CME leaving the Sun on the west side A, B-
  Frontal structure V- Void, C-core

V
5
What are CMEs?

• Large-scale magnetized plasma ejected from the
  Sun (part of the corona is expelled with its
  magnetic field)
• Propagate into the interplanetary medium and
  impact planets in the solar system
• The driving forces are not well understood, but
  related to solar magnetic fields and help push
  coronal material out of solar gravitational
  potential well.

6
Filament Eruption in EUV
Click on the image to start movie

• SOHO/EIT (195 A) images presented as a movie.
• The NW-SE filament (seen in H-alpha in the
  previous slide) erupts and becomes the core of
  the white light CME
• Arcade formation follows the eruption

7
2000/09/12 CME on the Disk
Click on the image to start movie

• SOHO/LASCO C3 movie
• Partial halo event consistent with the southern
  location on the disk
• The bright core is the filament that was dark in
  the previous movie

8
H-alpha Filament Eruption
9
He 10830
10
Three-part structure before eruption (Yohkoh/SXT)

• Frontal (but sheared)
• Cavity hidden?
• Filament core

11
Energy Release Near Filament
12
H-alphaBefore and After Eruption
13
Reconnection-favoring Flux Emergence
14
Closed and Open Magnetic Regions on the Sun
Coronal Hole
Active Region
Filament
15
Examples of Closed Field Regions
Active region
TIL
Filament
H-alpha picture
SOHO/EIT image 195 A
16
Where do CMEs originate?

• CMEs originate from closed field regions
• - Active Regions
• - Filament regions
• - Combination of AR and Filament regions
• - Transequatorial interconnecting regions
  (Gopalswamy et al. 1999, solar wind 10)
• CMEs do not originate from coronal holes!
• - Filaments near coronal holes show a proclivity
  for eruption (Webb et al., 1978 Bhatnagar, 1996)

17
Why Study CMEs ?

• Apart from the underlying physics,
• Long-term geoeffects the severest of geomagnetic
  storms are due to CMEs (Tsurutani et al., 1990
  Gosling et al., 1990)
• Long-lasting SEP events originate from CME-driven
  shocks (Reames, 1995)
• Energetic Storm Particles (ESPs) are carried by
  CME-driven IP shocks
• ? Main player in the Sun-Earth connections

18
Halo CMEs Originate on disk
19
Halo CMEs Front- or Back-sided?
20
Early History of Sun-Earth Connection

• Carringtons (1860) flare of 1859 September 1 at
  11 20 UT from N20W15. geomagnetic storms 18
  hours later
• Carrington was skeptical (good scientist)
• Confirmed by Hodgson (1859)

Flare CME relationship
21
What is a Geomagnetic Storm?

• Earths magnetic field is disturbed.
• Measurements of horizontal component of Earths
  magnetic field show disturbance lasting for a few
  days
• This is a result of currents induced in Earths
  magnetosphere when CMEs impinge on Earth

22
Dst Network of Observatories
San Juan
Kakioka
Honolulu
Hermanus
http//swdcdb.kugi.kyoto-u.ac.jp/dst2/onDstindex.h
tml
23
Example of Dst index
Quiet Period

• Dst index for April 2000
• Major and minor storms

Main Phase
Recovery Phase
24
MC on 97/05/15

• High Magnetic Filed
• Field Rotation
• Low beta

25
Storms and the Sun

• Sabine (1852) noted that the frequencies of both
  of geomagnetic storms and sunspots followed the
  11-year cycle.
• Maunder (1904), Greaves and Newton (1928a,b)
  noted that great geomagnetic storms were
  associated with sunspot groups occupying a large
  area on the disk.
• ? Greaves and Newton even deduced a travel time
  of 1.5 days for solar disturbances from Sun to
  Earth.
• Lindemann (1919) suggested that transient plasma
  ejections from the Sun impacted on the
  geomagnetosphere to cause the storms
• Newton (1943) found significant correlation
  between flares observed since 1892 and subsequent
  geomagnetic storms.

26
Storms and the Sun

• CH associated small storms with gradual
  commencements, recurring over 27 day period
  (Feynman and Gu, 1986 Maunder, 1905, Chree
  Stagg, 1927 Hundhausen, 1977)
• Transient Sudden commencement and large, from
  closed field regions on the Sun. The SSCs were
  due to IP shocks (Joselyn and McIntosh, 1981).

27
Proton Shower from the Sun
Solar energetic particles (SEPs) are measured in
units of Particle flux units (pfu) 1 pfu 1
particle/cm2/s/sr Discovered by Forbush (1946)
SEPs can damage Space Electronics, Solar Cells,
and pose radiation hazard to astronauts who
space walk.
28
Plasma Clouds Shocks

• Existence of plasma cloud -- Lindemann (1919)
  transient ejections of plasma from the Sun caused
  geomagnetic storm -- Chapman Ferraro (1929)
• Forbush (1938) had invoked plasma clouds to
  explain decreases in cosmic ray intensity.
• Gold (1955) suggested that plasma ejections from
  the Sun must often drive shock waves in the
  interplanetary gas.
• The plasma clouds were thought to have turbulent
  magnetic fields (Morrison 1956), smooth fields
  connected to the Sun(Cocconi 1958), and
  disconnected plasmoids (Piddington et al., 1958).
  
• Parker (1961, ApJ, 133, 1014) used hydrodynamic
  calculations to show that a 4 MK flare explosion
  could drive a blast wave to Earth in 1-2 days
• IP shocks were detected in situ by spacecraft
  (Sonnett et al. 1964) and were found to be
  relatively common (Gosling, et al., 1968
  Hundhausen et al., 1970).
• Hundhausen and Gentry (1969) solved time
  dependent hydrodynamic equations for IP
  disturbances and found both piston-driven and
  blast wave solutions.
• Even before the detection of CMEs, the IP
  disturbances were estimated to have a mass of
  1016 g and an energy of 1032 erg. (Hundhausen,
  1970)

29
Driver Gas, SEPs

• Two-step decrease in cosmic ray intensity
  (Forbush decrease) was correctly interpreted by
  Barnden (1972) as due to the ejecta (driver gas)
  and the extended post-shock region (Hundhausens
  two-part structure).
• The driver gas had different properties than the
  normal solar wind enhanced He (Hirshberg et
  al., 1972), low electron and proton temperatures
  (Gosling et al., 1973).
• The driver gas had counterstreaming superthermal
  electrons suggesting closed field lines contained
  in the driver gas (Montgomery et al., 1974).
• Lin and Hudson (1972) provided observational
  support the energy of gt 10 keV electrons can be
  sufficient to provide energy and mass for shock
  waves.
• Hundhausen (1972a,b) noted imperfect correlation
  between flare energy and IP shocks.
• Flux-rope structure -- magnetic clouds (Burlaga,
  1981)

30
Youngs Two-Ribbon Flare
Young thought these were prominences on the solar
disk. Now we know that prominences appear dark on
the disk because they are cool ( 8000 K) and
absorb radiation from the surrounding corona
31
Flares

• Following Carringtons white light flare
  (Carrington, 1860 Hodgson, 1859), the first two-
  ribbon flare was obtained by Young on 28 Sept
  1870 (the drawing shows a large two-ribbon flare
  although he thought that they are prominences on
  the Sun.
• Hales invention of the spectroheliograph to
  image the Sun and spectrohelioscope to identify
  rapid time variability and his vision to
  distribute his new instrument to various parts of
  the world began the patrol observations that
  started accumulating data on flares since 1934.
• Dellinger (1937) Sudden ionospheric disturbances
  were associated with flares (em effect).
• How flares are related to CMEs is a topic of
  current research

32
Prominences

• Well established in the late 1800s Secchi
  (1872) had already classified the prominences
  into active and quiescent
• Speeds of 100s of km/s were observed from
  spectroscopic observations (Fenyi, 1892).
• Greaves and Newton (1928b) correctly suggested a
  relationship between prominence eruptions and
  geomagnetic storms, but Hale (1931) pointed out
  that they fell back and subsequently dismissed by
  Newton (1939) since PEs rarely attained escape
  velocity.
• In 1947, Payne-Scott discovered the type II radio
  bursts and suggested a connection to solar
  eruptions.
• Now we know that PEs are integral parts of CMEs
  (Munro et al., 1979)
• Two classes of CMEs based on Inverse Normal
  polarity filaments (Low and Zhang, 2002)?

33
To Summarize the brief history

• Mass Ejections known for a long time from Radio
  bursts, H-alpha observations (e.g, Payne-Scott et
  al., 1947)
• The concept of plasma ejection known to early
  solar terrestrial researchers (Lindeman, 1919
  Chapman Bartels, 1940 Morrison, 1954 Gold,
  1955)
• CMEs as we know today were discovered in white
  light pictures obtained by OSO-7 spacecraft
  (Tousey, 1973)
• OSO-7, Skylab, P78-1, SMM and SOHO missions from
  space, and MLSO from ground have accumulated data
  on thousands of CMEs
• CME properties are measured in situ by many
  spacecraft

34
A 19th Century CME in Eclipse Data Eddy (1974)
3 part Structure!
35
CME Detection

• White light Thomson scattering of photospheric
  light Need occulting disk to block photospheric
  light (million times brighter than the corona).
  Samples mass irrespective of T.
• Other wavelengths Near-surface (H-alpha, X-ray,
  EUV, Radio) and IP manifestations (Radio, white
  light, IPS, In situ). T, n, B dependent.
• Mostly thermal emission (continuum, line). In
  radio thermal and nonthermal emissions (trillion
  K brightness temperature possible)

36
CMEs are Very Common

• gt 5000 CMEs observed by SOHO/LASCO until end of
  2001
• 0.5 per day during solar minimum
• Many per day during max
• Catalog http//cdaw.gsfc.nasa.gov

37
Many CMEs on the Same Day
CMEs observed by SOHO/LASCO - CMEs in many
directions - Some halos -Solar cosmic rays -
(Stars and a planet too)
38
Thats all!
39
Coronagraphs
40
Sun-Earth Connection

• Recognized in the 19th century The frequencies
  both of geomagnetic storms and sunspots followed
  the 11-year cycle. (Sabine, 1852).
• Maunder (1904), Greaves and Newton (1928a,b)
  noted that great geomagnetic storms were
  associated with sunspot groups occupying a large
  area on the disk. Greaves and Newton even deduced
  a travel time of 1.5 days for solar disturbances
  from Sun to Earth.
• Description of a singular appearance in the Sun
  on September 1, 1859 Carrington 1860). The flare
  was from N20W15. Hodgson (1959), an amateur
  astronomer observing nearby, confirmed the
  observations. The flare was followed by a severe
  geomagnetic storms in 18 hr. More details in
  Eruptive Flares (Svestka Cliver, 1991)
• Fully established in the early 20th century
  (Solar flares and magnetic storms - Newton,
  1943) Found significant correlation between
  flares observed since 1892 and subsequent
  geomagnetic storms.
• Dellinger (1937) Sudden ionospheric disturbances
  were associated with flares (me effect)
• Forbush (1946) SEPs - Ground level enhancement
  (GLE) of cosmic rays was associated with flares
• Reconnection (Dungey, 1961)
• Prominence Eruptions (Petit, 1932)
• IP shocks (Sonnett et al.1964)
• Magnetic Clouds (Burlaga, 1981)
• Hundhausen, 1972 Shock plasma cloud
• SEPs (Kuni Sakurai) Cliver 1996
• ESP (Rao et al., 1967) JGR 72, 4325
• Green line transients DeMastus, 1971
• Eclipse observations (Eddy)
• Payne-Scott et al., 1947) type II bursts
• CMEs in the modern sense (OSO-7) Tousey, R., The
  solar corona, Adv. Space Res., 13,713, 1973.
  (December 14, 1971)