For years it was thought that solar flares were responsible for major interplanetary (IP) particle events and geomagnetic storms. However, recently we have seen an important paradigm shift such that now coronal mass ejections (CMEs), not flares, are considered the key causal link with solar activity. CMEs are vast structures of plasma and magnetic fields that are expelled from the sun into the heliosphere. We now know that CMEs drive all large geomagnetic storms and their attendant effects, such as auroral displays. Fast CMEs produce transient IP shocks which cause sudden commencements at Earth. The CME-related shocks also accelerate the solar energetic particle (SEP) events associated with major IP disturbances and with radiation hazards at Earth. This new paradigm has important implications for prioritizing the development of instruments and techniques to predict IP disturbances and geomagnetic activity.
CMEs are best viewed in white light from space-borne coronagraphs; these reveal that, even near the sun, the CME can dwarf the solar disk [see Figure 1]. The onset of CMEs can be associated with both flares and filament eruptions [e.g., Webb, 1992; Feynman and Hundhausen, 1994]. However, most of the energy is associated with the ejected mass and shock wave, not the flare, even in the most energetic cases. Many CMEs cannot be associated with any surface activity. Most flares occur independently of CMEs and even those accompanying CMEs are now considered a secondary consequence rather than a cause of CMEs [ Kahler, 1992; Gosling, 1993]. Indeed, the basic physics of the two phenomena may be quite different.
The following sections review the evidence supporting the new paradigm in terms of the development and evolution of CMEs near the sun, the signatures of their passage through the interplanetary medium, and how and why CMEs are geoeffective. (Geoeffective means that an agent can significantly perturb Earth's environment.) Finally, we discuss implications of this paradigm for forecasting IP disturbances and geomagnetic activity. This is not meant as a comprehensive review but rather a portrayal of recent work in the light of this new concept. For further details about CMEs see the following review papers by U.S. authors: Steinolfson [1991; 1992b], Kahler [1992], Webb [1992; 1993], Gosling [1993]; Low [1993], Dryer [1994] and Hundhausen [1995].