What does this “web on the surface of the Sun” image reflect? What does 789 nm show us?

What is this web on the surface of the Sun? has got me thinking.

1. This is probably not a normal color photo.
2. The cooler areas are really dark!

1. The NSO press release says the passband is 789 nm.
2. "Images have been processed to remove noise and enhance the visibility (contrast) of small-scale (magnetic) features while maintaining their shape. The movie frames have been smoothed to remove noise."

Question: What does this image at 789 nm show us? Are we looking at changes in blackbody radiation or is there a spectral feature that tracks something more specific? Does intensity in the image actually track temperature directly, or reflect it in a more subtle way?

From The Universe Today's This is the Highest Resolution Image Ever Taken of the Surface of the Sun

• The NSO press release says the passband is 789 nm. – Mike G Feb 1 '20 at 4:09
• Note that sunspots are also quite dark, even when you look directly via a solar-filtered telescope. – Ruslan Feb 1 '20 at 7:50
• @MikeG excellent, so if we had a linear gray-scale image we could make a rough, approximate map of temperature by reversing Planck's law assuming the wavelength was chosen to reflect blackbody radiation rather than some temperature-dependent spectral feature. – uhoh Feb 1 '20 at 7:50
• I think the color themselves nothing. Their brightness yes. Ruslan is correct but our eyes adapt. I always wonder if this kind of solar pics are manipulated for contrast. – Alchimista Feb 1 '20 at 10:00
• @Alchimista The NSO press release that Mike G linked to says, "Images have been processed to remove noise and enhance the visibility (contrast) of small-scale (magnetic) features while maintaining their shape. The movie frames have been smoothed to remove noise." – Peter Erwin Feb 1 '20 at 17:22

Well let me take a stab at it. The line in question is said to be a probe of an Fe XI line, that is iron atoms with 10 electrons removed.

You do not get such ions in the solar photosphere, it is far too cool; the radiation from the photosphere is probably a pseudo-continuum at that wavelength.

However, much hotter material in the chromosphere and corona may contain Fe XI ions. Plasma in these structures could absorb light from the underlying photosphere, if it were dense enough, or more likely, if you point your camera above the limb of the photosphere, you could see optically thin coronal structures emitting light at this wavelength.

Why is this important? Normally you would need to use EUV or X-ray emission to probe the coronal plasma, but the imaging quality is not so good. You can do much better at optical wavelengths, but there are precious few diagnostic lines that can be used.

Edit: Indeed this presentation on DKIST coronal diagnostics appears to confirm this hunch and also mentions the additional possibilities that polarimetry of optical/IR spectral lines offer in investigating coronal magnetic structures. The Fe XI line in question is sensitive to Zeeman splitting, offering the opportunity to prove the strength and direction of magnetic fields where the line is formed. The Zeeman effect is proportional to the square of the wavelength, so the more usual X-ray and EUV diagnostics just can't do that.

On p.2 of that presentation it clearly states that these kinds of measurements are limited to coronagraphic images taken off the limb of the Sun. Pointing at the photosphere will not yield useful information for these diagnostics, because the weak light from the chromosphere and corona is swamped by the normal photospheric emission.

Sextus Empiricus has pointed to a press release, which although unclear in its exact meaning, implies that the dark features around the granulation may offer sufficient contrast to see the much thinner and hotter chromospheric gas directly above, producing bright points in the dark lanes. Observing these through different polarising filters could then reveal details about the structure and strength of the magnetic field.

However, on further research, these photospheric bright points are nothing of the sort. They are concentrated magnetic flux tubes that allow a deeper (and therefore hotter and brighter) view of the Sun. The typical temperatures of the deeper material is still only around $$10^4$$ K (e.g. Shelyag et al. 2010) and nowhere near enough to excite Fe XI.

The picture below, which comes from this site, referred to by Sextus Empiricus, shows the situation. A bundle of magnetic flux "hollows out" a passage further into the solar interior and the light we see comes from deeper, hotter, brighter regions. Nothing to do with coronal emission.

I arrive at the conclusion that this image (which was taken for science verification purposes) was just using the Fe XI filter as a narrowband filter. Everything we are seeing in the image is essentially continuum from the photosphere at temperatures between 4000K and $$\sim 10^4$$K. The contrast therefore just arises from the differing monochromatic intensity of material at different temperatures.

• This press release with this image shows the point of the Fe XI 789 nm filter well. They may be better references than the pdf. lso here dkist.nso.edu/node/319 – Sextus Empiricus Apr 9 '20 at 8:23
• @SextusEmpiricus neither of those links mentions Fe XI ? – ProfRob Apr 9 '20 at 9:08
• you are right they do not mention that the filter relates to the emission from Fe XI ions. But they do explain the coronal features and why they relate to the convection cells. In the particular image you can see the details of the flux tubes inside some of the darker parts in between the cells. The use of the filter is to observe those detailed features. To image the convective cells from the OP's question you do not need the 789.2 nm filter (and also that granular structure of convective cells had already been imaged decades ago and is less groundbreaking) – Sextus Empiricus Apr 9 '20 at 10:16

The Sun is pretty much a blackbody for every purpouse except when looking at it with a rather precise spectrometer.

Then again, it is not a constant temperature blackbody. The brightness of these images directly translates to some temperature in the corresponding region of the photosphere. The most bright of them are somewhere 6000K, the darkest pixels are, say, 4000K.

The color of the published images and videos is completely artificial and choosen to look "sunny". These 789nm are red in reality, near the red limit of human vision. Then again, they didn't say how wide is the passband of the filter.

• This web page describing the Visible Broadband Imager (VBI) used with DKIST shows that the wavelength band at 789.186 nm has a FWHM of 0.356 nm in width. This central wavelength is also identified as a spectral line of highly ionized iron (Fe IX). Since it is the largest wavelength available with the camera, it was likely chosen for maximum detail using the telescopes adaptive optics to help reduce degradation due to atmospheric seeing. – amateurAstro Feb 2 '20 at 22:42
• -1 because this seems more like a guess rather than a proper, supported Stack Exchange answer. A narrow filter with a passband centered on a spectral emission line is pretty much the same thing as "looking at it with a rather precise (imaging) spectrometer". – uhoh Feb 3 '20 at 0:33
• @uhoh Not exactly; the difference is that in an imaging spectrometer, you have the intensity at multiple wavelengths, whereas in this image, you only see the intensity at 789.186 nm, corresponding to the Fe XI line. I don't think the exact shape/width matters, since this is a narrowband centered on a line that dominates the continuum. – pela Feb 3 '20 at 11:04
• That said, this answer could benefit from an explanation of what the Fe XI line shows us (and possibly how the brightness can be translated to temperature). – pela Feb 3 '20 at 11:09
• @pela okay, an "imaging monochromator" ;-) – uhoh Feb 3 '20 at 11:40