At last, the EUV Spectrum

From RHESSI Wiki

Jump to: navigation, search

Number: 145
1st Author: Hugh Hudson
2nd Author:
Published: 7 February 2011
Next Nugget: M is for Magnifique Part Deux
Previous Nugget: Black and White Flares
List all



We can detect stellar flares easily, in spite of the competing brightness of the star. Until recently, though, we could not do that for solar flares: only through imaging, as in the original Carrington flare, did the excess emission from the flare actually become detectable. Partly this reflects the much greater surface brightness of the Sun, as compared with that of the usualy faint dMe flare stars, and partly no doubt it reflects the relative ease with which ground-based astronomers can do precise photometry on a point source. Now (at last!) we have clean observations of the all-important impulsive phase of a solar flare.

New "Sun-as-a-star" observations of solar visible and EUV emission have revolutionized our understanding of the energetics of solar flares. Figure 1 of an earlier Nugget shows the breakthrough observations (from the SORCE satellite) of the flare SOL2003-10-28, a GOES X17.2 "superflare." Viewing the Sun as a star (i.e., measuring the total solar irradiance or solar constant), the SORCE radiometer found a trifling 0.03% increase for this huge flare. No wonder it took a century and a half of technical development beyond Carrington to get to this point.

In this Nugget we describe a major step towards identifying the wavelengths of the SORCE radiation, which (as many suspected) has a close relationship to the white-light flare mechanism. The new insights come from the EVE instrument, which we have already introduced in an earlier Nugget.

A well-observed gamma-ray flare

The newly-launched Solar Dynamics Observatory has a Sun-as-a-star EUV spectroscopy instrument, EVE, led by Tom Woods. This gives solar astronomers, for the first time, continuous spectra (at 0.1 nm resolution and 10 s sampling) of basically the entire EUV spectrum (6.5-105 nm, plus Lyman-alpha). The instrument has other features, including a set of very sensitive broad-band EUV radiometers. Figure 1 shows the passbands of these photometers, plus their responses to the flare we described in the previous Nugget: a flare exhibiting white-light continuum and gamma-ray emission. The flare we report, SOL2010-06-12, stands out thus far in Solar Cycle 24: a white-light flare, a gamma-ray flare, and a flare brighter at 80 GHz than even the memorable flare SOL2005-01-20.

Figure 1: Left, the four passbands of the EUV/ESP instrument; right, their responses during the flare SOL2010-06-12. Note the varying degrees of impulsive (early) and gradual (late) as the many different spectral lines contribute their counts to these broad-band photometric channels. The passband at lower right resembles the GOES long-wavelength channel: a broad band centered at 4.4 nm.

This figure only gives a glimpse of the data, which included many details of individual spectral lines and their properties. Note that even solar flares have excellent observability in the EUV, because the photosphere has too low a temperature to emit much at these short wavelengths.

UV spectral synthesis

We would like to make a spectrum that shows just the impulsive phase of the flare, in order to understand the flare energy seen in the TSI signal. Figure 2 (left) shows when this phase happened for this event, by appeal to RHESSI hard X-ray emissions above 100 keV. We have used this as a guide and generated the broad-band spectrum in Figure 2 (right). The spectrum figure shows the white-light flare continuum schmatically a 10,000 K blackbody; the slightly diagonal dashed line in the figure represents the total signal from the photosphere - much larger in area, but at a somewhat lower temperature. The EUV points lie well below the white-light flare continuum in this plot of the energy distribution. We believe that this spectrum provides the best information yet regarding the spectral distribution of the SORCE observation of the TSI signal described way back in 2004 (see Nugget No. 10).

Figure 2: Left, RHESSI 100-200 keV emission (thin line) compared with He II 30.4 nm radiation; right, a broad-band spectrum of just the impulsive phase. The smooth line roughly describes the white-light flare continuum, the points the EUV background from EVE, and the diamonds, the EVE broad-band points. The dashed line shows a blackbody corresponding to the photosphere.

The spectrum (Figure 2, right) shows a spectral energy distribution: we have plotted νfν (the same thing as λfλ). Such a plot has equal areas for equal energies per decade (or per octave), so that one can see at a glance which wavelengths dominate. In this case the EUV and X-ray bands do not dominate; instead we have narrowed the principal energy contribution to the near UV. EVE has data that cover this range as well (though not for this flare), and we look forward eagerly to analyzing them.


This atypical single flare hardly lets us draw sweeping conclusions, except for the obvious one that the EVE instrument can indeed make brilliant observations of the Sun as a star, and can characterize the EUV spectrum of the impulsive phase of flares. The importance of this capability lies in the information we can glean about the fundamentally important energetic processes of a flare - its luminosity and its eruption of mass, both into the corona and further out into the solar wind as a coronal mass ejection.


[1] Extreme Ultraviolet Variability Experiment (EVE) on the Solar Dynamics Observatory (SDO): Overview of Science Objectives, Instrument Design, Data Products, and Model Developments


Thanks to Stephen White, Phil Chamberlin, and Ludwig Klein for input

Personal tools