Instantaneous Flare Properties
From RHESSI Wiki
|1st Author:||Hugh Hudson|
|Published:||February 10, 2014|
|Previous Nugget:||Reflection of Coronal Global Waves|
To not much fanfare, Pat Bornmann described (in 1990) how to think about the significance of background corrections in GOES soft X-ray data [Ref. 1]. One essential realization involved in her development is that the increase of soft X-ray fluxes might come either from a new injection of hot plasma, or from a physical change in plasma already contributing to the steady background. In this Nugget we descrbe a related method, that of the analysis of the instantaneous increment of the GOES signals. This essentially is a study of the time derivative of the data.
In the case of solar flares, measurements (such as those of GOES or RHESSI) can often be made on quite short time scales (the Sun is bright). Accordingly it might be safe to assume that changes in physical parameters do not have time to vary significantly, and that a "constant density" limit prevails [Ref. 2]. In other words, the structure must appear frozen in place on short enough time scales. In such a case we can do a standard GOES temperature and emission-measure analysis on the increments of the two channels, an approximation to their derivatives, and get a different view of the nature of the change. This technique seems quite general but as yet there is not much literature exploiting it.
We illustrate the technique first with an amazingly impulsive GOES M1-class flare, SOL2013-01-05T09. This event occurred in an active region just beyond the E limb of the Sun, and so its X-ray time profiles may reflect this circumstance. A recent paper [Ref. 3] describes the event as a "failed eruption," which is kind of the polar opposite of the "breakout model", in that the same sort of reconnection is assumed to happen in the same way, but not to cause an eruption. How can both views be right?
Because this event was so impulsive, we used it to develop the "GOES incremental" view of its X-ray time history, as described here.
"The GOES incremental" view
The method is simple. We just use each point in the GOES time history, which in the most recent data are at 2-sec spacing, as a background reference for the next point. It is a running difference; we use the standard GOES software to do the fit to the increments from the two channels. The temperature and emission measure that result, point by point, are those of the increment in the 2-sec interval. Figure 2 shows two examples, one from the very unusual and very impulsive SOL2013-01-05T09:31, and one from a different kind of event.
Both of the flares in Figure 2 point to the same conclusions. The incremental temperatures tend to exceed the integral temperatures during the rise phases of each burst, and they may be below the integral temperatures at other times, even when the event is developing. The peak temperatures approach 40 MK, substantially beyond usual flare temperatures, which strongly implies non-thermal contributions considering the time scales involved.
The interesting thing here is that even low-temperature increments can add to high-temperature accumulation. This remark seems to cry out for modelization, since it seems so topsy-turvy intuitively.
The extension of Bornmann's methods that we have explored here might be important in the exploration of transient astronomical phenomena observed in other ways. The time derivative of something, after all, is not the same thing as the thing itself. There is not much (if any) solar literature on this kind of analysis.
New high-precision instantaneous and frequently-sampled data may make thus new information available; since the Sun is generally so bright, it can support timing observations of (astronomically speaking) incredible sampling and precision. In the case we have studied here, the GOES 2-sec data, we are not really approaching the precision necessary for a good estimate of the Sun-as-a-star X-ray temperature; in two seconds, at an Alfven speed of 104 km/s, as might be found in a flare, suddenly-heated material might expand over many gravitational scale heights in the chromosphere. This would confuse the Syrovatskii "constant density" idea. But it is still there for the taking, and I suggest that this would be a good prospect for a Cubesat experiment.
-  Limits to Derived Flare Properties using Estimates for the Background Fluxes - Examples from GOES