Understanding HMI pseudocontinuum in white-light flares

From RHESSI Wiki

Revision as of 17:55, 22 August 2018 by Schriste (Talk | contribs)
Jump to: navigation, search

Number: 324
1st Author: Michal Švanda
2nd Author: et al.
Published: 28 May 2018
Next Nugget: A remarkable, but confused, coronal hard X-ray source
Previous Nugget: To Beam or not to beam - that is (still) the question, To beam or not to beam - that is (still) the question
List all



The most violent phenomena of solar activity are termed flares. Aside from the physical processes underlying them, these flares contain sources of electromagnetic radiation at all wavelengths from radio waves to gamma-rays. White-light flares (flares with emission in the visible continuum) are probably not a rare phenomenon. The enhancement relative to the photospheric brightness can be low, typically a few tens of percent, and it can be challenging to detect even in an X-class event. The energy contained in the white-light flare continuum represents a large fraction of the total flare energy.

These infrequent white-light flares have appeared to observers via many different instruments in the past. In order to properly study this phenomenon, continuous high-resolution data are definitely desireable. A synoptic experiment such as the HMI instrument on the SDO satellite therefore represents the holy grail of white-light flare observation. The HMI data have now been used in several studies of white-light flares. These data consist of sequences of exposures at different polarizations across six narrow wavelength bands capturing an Fe I line at 617.3 nm, an absorption line formed in the [photosphere]. Students of white-light flares in these data usually employ the standard data product called Ic, an estimate of the continuum level based upon this coarsely observed line profile. This is available every 45 s, a coarse sampling of the actual flare variations.

Comparison of HMI and Hinode observations

We have studied an X9.3-class flare that occurred on September 6, 2017 (SOL2017-09-06T11:53). We take advantage of the observations taken by the Solar Optical Telescope (SOT) aboard the Hinode satellite. For the first time, Hinode recorded detailed emission profiles during such an event, while HMI observed in its regular mode. We could use of the full set of HMI data across its polarization and wavelength settings for an exact comparison. Hinode/SOT observed the Stokes profiles of the two Fe I lines at 630.15 nm and 630.25 nm in a raster scan covering the flare ribbons. The two sets of observations were aligned both in space and time to ensure that we compared the profiles observed at the same pixel (see Fig. 1). Both HMI and SOT observe the solar photosphere, yet in different spectral lines. Given the similar formation height of the observed spectral lines, we use a complete model atmosphere derived from the Hinode observations to synthesise the Stokes profiles of the HMI line. This essentially calibrates the coarse HMI data product against a much more precise Hinode/SOT observation.

Figure 1: The co-aligned fields of view. Left: Hinode/SOT spectral scan displayed at continuum spectral point of intensity profile. The scanning was performed in the horizontal direction indicated by an additional time axis on the top. Right: the co-aligned pseudoscan from HMI. The red dashed lines indicate the transition between the HMI frames in different times (that are also given in red letters). Note that the outermost wavelength filter position was used in this plot. The times represent UT September 6, 2017.


Our analysis covered 5475 pixels, including quiet Sun, sunspot penumbra and umbra, and the region showing emission profiles. In the quiet-Sun regions, the synthetic profiles of the Fe I 617.3 nm line and the HMI observations match quite well on average. The HMI spectral point outermost from the line centre is located already in the line continuum and thus may be used as a measurement of the nearby continuum level. In the umbral pixels the intensities of HMI spectral points are systematically larger than predicted by the model, which is likely an effect of the scattered light in the instrument. The line profiles in the umbra are very wide in general and none of the HMI spectral points lie in the line continuum; they all sample the line core. Additionally, the spectral line profiles may have a quite complicated structure, therefore a simple algorithm often cannot determine the continuum level accurately. Depending on the real line profile in the pixel and the sampling points, the algorithm may either overestimate or underestimate the real continuum value. In the penumbral pixels, the line profiles are broader than in the quiet Sun, but narrower than in the umbra. When the emission profiles are present, the spectral line often has a very complicated shape, and with only HMI's six spectral bins, the true nature of the profile cannot be resolved. Since most of the points in the emission are either in the umbra or in the light bridge, the line is broad here and the fitting is worse. Therefore, in general, the Ic values are unreliable in the areas with the line emission. The modelled continuum based on Hinode observations generally shows an enhancement but its value is much lower than the values indicated by Ic product. The relative error of the HMI pseudocontinuum may reach tens of percents.


We directly compared data products from observations by Hinode/SOT and HMI recorded during a very strong X-class flare. Our aim was to learn about the Ic product available from HMI, widely taken to characterize the photospheric continuum brightness. We show that in the quiet-Sun regions the match between the Stokes profiles modelled from Hinode/SOT SP data and detected by HMI is acceptable and that the value can serve as an indicator or a proxy for the continuum intensity. In magnetised regions the situation is far less satisfactory, and the value of Ic may be strongly biased. This is due to the fact that all six HMI spectral sampling points are located within the line core broadened by the magnetic field. Therefore, the Ic value is estimated using a model that is too simple. The X9.3 flare SOL2017-09-06 also showed emission profiles in the flare ribbons. The HMI pseudocontinuum in the these pixels is again strongly biased. We stress that without the knowledge of the true line profile in any given pixel, it would not be possible to assess the biases properly. The example in Fig. 2 shows a case when a profile with a strong emission is sampled by the six HMI points in such a way that one would consider this line to be a simple absorption line. Our model of the four Stokes line profiles matches the observed ones reasonably well, but the reported Ic value is substantially off. While Ic may be used as an indicator for the propagation of flare ribbons, it must be used with caution in any kind of photometric or energetic studies of flares.

Figure 1: The four Stokes profiles for a representative pixel with line emission. The solid black line represents the synthetised 617.3 nm spectral line based on the atmosphere derived from inversions of Hinode data, smoothed with a Gaussian of FWHM=7.6 pm (to approximate the HMI transmission filters). The red diamonds indicate the spectral points derived from HMI data series and the vertical red lines mark the estimated noise level. The red triangle denotes the Ic HMI pseudocontinuum.


J. Jurčák, J. Kašparová, and L. Kleint all contributed to the published paper (Ref. [1]) and to this Nugget.


[1] "Understanding the HMI pseudocontinuum in white-light solar flares"

Personal tools