Categories of Prominence Eruption or Disappearance

qs084.martin01
Posted:  27-Nov-94
Updated: 29-Jun-95, 26-Apr-96
Events specified: N/A


Proposers: S.F. Martin, R.C. Canfield, K. Shibata, A. McAllister, K. Reardon, H. Tonooka and interested contributors

Using prominence data from the H-alpha limb filtergraph of the Mees Solar Observatory, we have verified the existence of at least three categories of prominence eruption or disappearance. The most common category is the erupting prominence in which some of the filament mass is injected into the high corona and is thought to escape the sun while the remainder of the mass falls back to the chromosphere. The second category is the `quasi-eruptive' in which the prominence begins to bodily rise, similar to a complete eruption, but all of its observable mass in H-alpha flows to the chromosphere near the apparent ends of the prominence. The third category are those prominences in which the mass drains out of the prominence without evidence of either slow ascent or eruption. The latter two categories are thought to be relatively rare. Information on these possibly different types of prominence disappearance is needed to ascertain if a single model can account for all prominence eruptions or disappearance or whether multiple models wil be required.

We propose to continue identifying the types of X-rays events
that do or do not accompany these categories of disappearing
prominences.  Our first goal is to determine if specific X-ray
characteristics systematically occur with the differing
categories of erupting or disappearing filaments. Our second goal is to understand the complete spectrum of erupting and disappearing filaments and accompanying X-ray events which need to be modeled.

Data to be used: SXT full sun images

Update 26-Apr-96

The study of Martin and McAllister on the relationship of the skew of coronal X-Ray arcades to the chirality (handedness of filaments (dextral or sinistral) has been expanded to include more dynamic arcade events. The overall results are the same but the sample (36) is now large enough that it includes exceptions to the hemispheric pattern in which left-skewed arcades are dominant in the northern hemisphere and right-skewed arcades are dominant in the southern hemisphere.

The Martin and McAllister results have broad application to the categorization of filaments during their eruptive phase and to associated X- Ray events, CMEs, and subsequent interplanetary clouds. In a follow-on paper in progress, McAllister and Martin suggest that the skew of coronal arcades can be used to predict the sense of helicity in associated X-ray features, erupting filaments, CMEs, and interplanetary clouds. Work has begun on the verification of this technique.

Recently, Tonooka has found two events that were classified as eruptives and now suggests that these events might be quasi-eruptives because of evidence of coronal loops forming above the prominence mass. McAllister and Martin have studied a filament eruption observed in X-rays and H- alpha against the disk. The event is believed to be a quasi-eruptive because no section of its mass abruptly disappears; the filament mass remains extraordinarily active as the flare loops apparently develop above it. Eventually the mass slowly disappears, presumably by falling back to the disk. McAllister has also found another event at the limb which is apparently another example of a quasi-eruptive initially thought to be a complete eruptive. There is accumulating evidence that quasi-eruptives might not be as rare as previously supposed. An next important task is to learn whether any of these quasi-erupting filaments were accompanied by CMEs or associated with interplanetary events.

Update 29-Jun-95

A study of the orientation of coronal arcades relative to sinistral and dextral filaments and filament channels was conducted by S. Martin and A. McAllister. The study included both quiescent arcades and dynamic arcades. The coronal arcades were defined as left-skewed and right-skewed relative to the filaments and/or polarity inversions beneath the arcades. These two categories differentiate between arcades having a magnetic field component either parallel or antiparallel to the long axis of the filament aand/or polarity inversion (in addition to the remaining component orhtogonal to the polarity inversion).

For a sample of 33 quiescent arcades and 11 dynamic arcades, it was found that only left-skewed arcades exist above dextral filaments or filament channels and only right-skewed arcades exist over sinistral filaments or filament channels. The global magnetic pattern was verified for filaments and filament channels (Martin, Bilimoria and Tracadas 1994) and thus is shown by this study to apply as well to coronal loop systems overlying filament channels and filaments. 14 of 17 dextral filaments and overlying left-skewed arcades were in the northern hemisphere and 13 of 16 sinistral filaments and their overlying right-skewed arcades were in the southern hemipshere.

A paper is being written on this study. The results were presented at the IAU Colloquium No. 153 in Japan in May 1995 and at the Solar Physics Division meeting of the AAS in Memphis, Tennessee, USA in June 1995.

A superposition of an SXT image and an image of the H-alpha filament for the 12-17 May 1992 data set was made by A. McAllister and an analyses of these data has been initiated by A. McAllister and S. Martin. The study of this filament and associated X-ray structures is of particular interest. The overlying arcade is very tenuous in this example but the superposition of the Ha images over the X-ray image shows a close fitting X-ray cloak which covers most of the top and sides of the H-alpha filament. The X-ray cloak is longer that the H-alpha filament and is nearly parallel to the long axis of the filament.

Data is being assembled and reduced for a study of the filament and filament channel formation observed in July 1994. An example of the formation of a filament channel and filament observed at the Ottawa River Solar Observatory in July 1979 is also being studied by V. Gaizauskas and J. Zirker. This earlier example provides an excellent comparison for the July 1994 example.