Significant new results on white light CMEs derived from analyses of
images from the Solar Maximum Mission (SMM) orbiting coronagraph,
the Mauna Loa Solar Observatory (MLSO) K-coronameter, and the
zodiacal light photometers on the Helios spacecraft have been
published. These results add to earlier studies of data from the
coronagraphs on Skylab and on the P78-1 spacecraft (called SOLWIND).
It is now clear that there is a large range in the basic properties
of CMEs. Their speeds, accelerations, masses and energies range over
2--3 orders of magnitude. The masses and kinetic energies of SOLWIND
CMEs extend from 10
to 5 x 10
kg and 10
to 6 x
10
j, respectively [ Howard et al., 1985], and some SMM
CMEs exceed these values [ Hundhausen, 1994]. CME occurrence
rates vary by an order of magnitude over the solar cycle [ Webb
and Howard, 1994]. CME widths range from 10 to
120
, exceeding by factors of 3--10 the sizes of
flares and active regions.
In white light images CMEs often appear to have a bright leading looplike structure followed by a dark cavity and a bright core of denser material, suggesting the eruption of a preevent prominence, its overlying coronal cavity, and the ambient corona. However, the frequency of occurrence of these structures differs from one coronagraph to another; for example, the fraction of looplike CMEs ranged from 1% of SOLWIND CMEs to nearly half of the SMM CMEs. Two-thirds of the SMM CMEs also contained bright cores. It is not obvious that these differences can be explained merely by solar cycle effects. The leading CME structures are likely the skyplane projections of three-dimensional structures such as arcades [ Steinolfson, 1992a] or shells [ Hundhausen, 1994].
From what kind of structures do CMEs emanate? Recent studies
confirm that CMEs arise from large-scale, closed structures, most
of which (
75%) are preexisting coronal streamers [
Hundhausen, 1993; Webb et al., 1994]. This is not
consistent with the suggestion of Hewish and coworkers [e.g.,
Hewish and Bravo, 1986] that CMEs arise in open field
regions (coronal holes) [ Harrison, 1990; Hundhausen,
1993]. The temporal and latitudinal distributions of CMEs are
similar to those of streamers and prominences, being confined to
low latitudes about the current sheet near cycle minimum and
becoming distributed over all latitudes near maximum [
Hundhausen, 1993]. This evolution is very different from that of
active regions, flares or sunspots. Many energetic CMEs are
actually the disruption of a preexisting streamer, which
increases in brightness and size for days before erupting as a
CME. Afterwards the streamer and CME are gone giving the
appearance of a ``bugle'' on white light synoptic maps [
Hundhausen, 1993]. Often a thin ray appears in the location of
the streamer, and may be a current sheet rising from a newly
reformed streamer. Feynman and Martin [1995] find that
major erupting filaments are strongly associated with emerging
magnetic flux oriented so as to favor reconnection of field
lines. Since filament eruptions are associated with CMEs and
streamers, this suggests that streamers can be destablized by
emerging flux. SMM CMEs have also been found to be associated
with large-scale evolving patterns of existing surface flux,
delineated by polarity inversion lines [ Webb et al., 1994].
What are the kinematical properties of CMEs? The speeds of the
leading loops of SMM CMEs ranged from 20 to 1100 km/s [
Hundhausen et al., 1994]. The average speeds of SOLWIND CMEs
were much higher near solar maximum than minimum [ Howard et
al., 1986], but SMM CME speeds were not solar-cycle dependent.
Significant progress has been made in simulating the coronal
response to the passage of a CME [reviewed by Steinolfson,
1992b and Hundhausen, 1994]. Fast magnetohydrodynamic
shocks apparently driven by fast CMEs are the most common type in
the inner heliosphere. However, in the lower corona the speeds of
typical CME outer loops are higher than the sound speed but less
than the Alfven speed (for SMM averaging 445 km/s) and the
trajectories are consistent with constant speed. Numerical
simulations of the disruption of model streamers indicates that
slow, intermediate and fast mode shocks should form ahead of CMEs
with speeds of 200--300, 300--900, and 
900 km/s,
respectively. Thus, slow and intermediate mode shocks might be
associated with most CMEs, and some CMEs do exhibit the predicted
flattened fronts. The existence of fast mode shocks in the corona
is strongly supported by the observation of rapidly drifting
radio bursts and their association with fast (
400 km/s)
CMEs [e.g., Kahler, 1992]. However, it is unclear whether
they compress sufficient material to be detected optically as bow
waves in front of CMEs.
In previous years much theoretical work on CMEs was focussed on reconnection of the magnetic fields which close after the CME has erupted. The recent models of this process describe the late phase reasonably well [cf., Svestka and Cliver, 1992; Webb et al., 1994]. In such models field lines stretched open during the eruption of a prominence and CME reconnect near the surface to form a magnetic loop system, the long-duration/ two-ribbon flare. New observational results generally support this concept. Kahler and Hundhausen [1992] found that the bright structures following many SMM CMEs are streamers probably newly formed by reconnection. Observations from the Japanese Yohkoh spacecraft and from MLSO of the reformation of a giant helmet streamer also provide strong evidence of reconnection following CMEs [ Hiei et al., 1993]. Another Yohkoh observation suggests rapid reconnection following the ejection of a plasmoid associated with a two-ribbon flare [see Webb et al., 1994]. This observed structure was found to be in excellent agreement with field lines derived from a circuit model, lending further support for the two-ribbon flare scenario.
Most of the models intended to describe the origin and propulsion of CMEs are not sufficiently developed to compare with observations. Most such models involve force-free equilibria which cannot realistically describe the complex evolution of the pressure, magnetic and gravitational forces acting on a magnetically closed coronal structure [e.g., Hundhausen, 1994]. The class of models which require a thermal or pressure pulse (i.e., flare) as driver no longer seem viable [cf., Dryer, 1994; Webb et al., 1994]. For instance, such models are not consistent with CMEs which exhibit significant accelerations over large distances. Recently there has been intensive work on the origin of CMEs based on the slow evolution of particular coronal structures through metastable states or sequences of stable equilibria until the stability or equilibrium breaks down, resulting in the mass ejection and opening of the field. Steinolfson [1991], Low [1993] and Dryer [1994] review such analytic models and numerical simulations. Causes of the evolution of these coronal structures, especially streamer configurations, include the emergence of magnetic flux [ Steinolfson, 1992a], the dynamical evolution of arcades [ Steinolfson, 1991; Mikic and Linker, 1994], and the shear of field lines across inversion lines [ Steinolfson, 1991; Wolfson and Low, 1992; Mikic and Linker, 1994]. However, no strong consensus has yet emerged.