Coronal mass ejections (CMEs) are an integral part of the "standard model" for flares. Yet flares without CMEs are not uncommon, and such flares can sometimes be extremely energetic, reaching GOES X class in some well-documented cases. What's going on in these types of flares? Why don't these eruptions expel CMEs? This is the question to be addressed by this subgroup.
As with many phenomena on the Sun, the answer to why flares-without-CMEs (non-eruptive flares) occur is probably a consequence of the magnetic configuration; the same basic process that drives eruptive flares may be going on in the non-eruptive flares, but in the non-eruptive case the overlying magnetic fields are very strong and end up confining the eruption, as discussed by, e.g. Moore et al. (ApJ 552, 833, 2001). We will test this idea and consider other ideas for the cause of non-eruptive flares in this subgroup.
As preliminary work on this topic, we (mainly Ron Moore) have put together a list of about 20 candidate flares for analysis, almost all of them from 2002. These flares were selected because of their strong but short-duration GOES soft X-ray profiles, typically lasting less than 2 hours; previous work suggests that such flares tend to not expel CMEs (Sheeley et al. 1983, etc.). Inspection of the SOHO/LASCO CME catalog indeed shows that most (15) produce no CME, but some (5) do initiate CMEs. All of the flares occur in strong magnetic fields near sunspots and most are compact (brighten only a small part of the active region). From their magnetic setting determined from MDI magnetograms and EIT images, it appears that the non-CME flares are confined deep within strong closed magnetic fields, while the CME-producing flares either trigger neighboring sheared field to erupt into a CME, or are able to erupt out of the active region and directly drive a CME. Compact ejective flares of this last type can produce a large CME by driving the eruption of a large coronal loop such as a trans-equatorial loop.
So far we have used LASCO, EIT, and MDI data for our analysis. During the workshop we would like to combine these data with data from other sources, including RHESSI, H-alpha, and radio, in order to test the tentative conclusions noted above. We encourage participants with these or other data for events in the list to contribute to our analysis, and we invite all participants to take part in discussions.