Weekly Notes from the Yohkoh Soft X-Ray Telescope

(Week 21, 2002)

Science Nugget: May 24, 2002

Comparing coronal dimming at different temperatures


Many of us believe that "coronal dimming" is an important observational signature for understanding the origin of coronal mass ejections (CMEs). In fact, dimming has been taken up several times in these science nuggets (for example, see 1, 2, 3). However, we still do not know how the dimming maps to the CME as observed by a white-light coronagraph. A more basic question is "does the dimming really represent mass depletion?" One may argue that change in temperature can also lead to a decrease of signals. Therefore, it is important to make multi-wavelength observations of the dimming phenomenon. Such studies (e.g., paper by Zarro et al.) have been not very common. Here, we compare SXT and EIT data for some dimming events most likely associated with CMEs. Note that EIT is more susceptible than SXT to temperature effect, but it appears that most of the EIT dimming represents mass depletion due to CMEs.


Contrasting examples

The first event we take up here occured during solar minimum. In a previous nugget, an SXT difference image (including super-long exposures) was given to show dimming around the time of a CME. Here we replicate the figure made by PI of the SXT project (click the following thumbnail to enlarge.)

Note that the dimming region is much smaller than the angular extension of the CME. A movie of SXT images together with a plot of the SXT light curve indicates that dimming preceded what appears to be a filament eruption. Interestingly, the following figure shows that EIT failed to reproduce the dimming region except in the 284 channel. This implies that only high-temperature (> 2 MK) plasma was involved in the process.

We should say this pattern (dimming observed by SXT but not in the EIT 195 channel) is quite uncommon. The following example was reported in a previous nugget. Here the CME was most likely associated with a major flare which would have been X-class. The CME roughly encompassed an EIT dimming region as shown in this figure. The following figure shows the light curve of EIT dimming, which started 15 minutes before the CME was first detected by LASCO (as indicated in the vertical dotted line). The boxed area continued to be dimmed for hours only gradually recovering around 04 UT.

In the previous nugget, we reported that the EIT dimming region roughly matched a bright region in SXT. Further analysis reveals that at no instant was dimming seen in SXT data.

Disk events

Now we look at dimmings away from the limb. They are often referred to as transient coronal holes and sometimes appear as double. We first recall that in the 1997 April 7 event, the dimming regions were similar in SXT and EIT 195 data (see the Zarro paper, and also the following figure of SXT and EIT data).

But such a good match seems to be not always the case. For example, a similar event was observed by EIT on 2001 November 21, which also served as an example of global eruption from a S-shaped region (see a previous nugget). The following figure shows that only EIT data nicely caught dimming around the time of the CME (the dotted vertical line). This is because the SXT dimming tends to be of smaller amplitude and in this case the flare emission affected the regions of EIT dimming.

Trans-equatorial events

Some CMEs originate from the eruption of large-scale loops including those that cross the equator. These are characteristically flare associated and the flare is located near one end of the large-scale loops (see the paper by Khan and Hudson, but sorry no abstract at ADS). Interestingly, the disappearance of the trans-equatorial loops was seen only in SXT difference images. In EIT, different areas dimmed, and the SXT dimming region appeared bright. See the following figure.

This pattern indicates that the SXT dimming may have been due to cooling of the loops, which then appeared in EIT, but we need more quantitative analysis. In addition, we have to mention that we could not find a better example than this one (1998 May 6, as well as May 8 and May 9) of the disappearance of trans-equatorial loops in association with CMEs, when we analyzed data for all the X-class flares in this cycle. This will be presented in a later nugget.


Consistent with one's intuition, the dimming signatures in SXT data can be subtle, and are not always easy to get. Sometimes flare emission occupies the same line of sight as the pixels that may have experienced dimming. At other times, the plasma contributing to a CME has temperature lower than SXT can see (but remember the 1996 October 5 event shown above as a counter example). There should be a contribution of increasing or decreasing temperature to the EIT dimming, but the good correlation with CMEs indicated that mass depletion is a dominant factor. The appearance of dimming in SXT data due to CMEs may depend on the intensity and location (including how far behind the limb) of the associated flare, and possibly the type of ejection if there is a variety. The plasma parameters of the CME with respect to the other part of the corona may have some solar cycle dependence.

[Topical index] -o- [Chronological index]

May 24, 2002

N. Nitta (NVN) (nitta@lmsal.com)