Summer solar visual and imaging challenge : "What is the color of sunspots ?"

[DanielBrousseau]

I tried with my ASI294 mono binned 2x2 with RVB filter form Astronomiks.
I calibrated each color the best I could naturally, as any planet.

There is some color on the penumbra but it's not any science for my.

[EdAstle]

Visually through my C9.25 (ND5 film), alessbonsai nailed what I saw:
viewtopic.php?p=397453#p397453

"Straw" penumbra exactly like that.

My eye saw nothing but black in the spot.

Not surprising considering my pupil was constricted. Intensity is proportional to the 4th power of temperature, so I've read.
My tiny eye hole stood no chance against that lowest chilly temp with inverse ^4 applied.

The color of sunspots will change as the sun ages. Final answer!
One day they will be brilliant white and then never exist again.

The day the sun explodes Lunt 100 DS will be cheap as chips on ebay. Yay! 8 minutes of..oh..where did it go. Pff Too late

[Montana]

This is hard!! how on earth do you set the surface to the right colour temp to start with?

Sun_131309_20_08_2023 by Alexandra Hart, on Flickr

Sun_130712_20_08_2023 by Alexandra Hart, on Flickr

Sun_131037_20_08_2023 by Alexandra Hart, on Flickr

Alexandra

[EdAstle]

There must be a palette of colours inside. A whole bunch of photons of different frequencies.
The RGB mix must fluctuate within.

Page 104 of the Solar Astronomy book asks "Do sunspots have color"?
The answer the book gives, in a nutshell, its "yes". ~3800K is the number given.

But you can't give a single answer - it's "beige" or "ruddy" or "sepia".

The human eye is...variable in calibration. The colour perceived will vary, and ones ability to nominate a named colour will vary on ones exposure to "colour names" as well as physiological differences.

For example, the colour "vellum"; not everyone knows that one, do they Alexandra.
Neither did I, for what it's worth

Because of this stupid bloody question I've now bought Antlia IR685 + IR850 to see colours I can't see but my Player One M-MAX can just to prove, well, even if there is no colour, there is. And my monitor will ultimately answer "it's not black". Thanks for that.

[Montana]

Rather than using automatic white balance in Photoshop, this time I tried again after some research. I used Levels and made sure I set the grey arrow in the middle of each red, blue and green histogram. I hope this is a more representative version of the colour of the Sun (no vibrance used).

2023-08-20 colour by Alexandra Hart, on Flickr

Alexandra

[DeepSolar64]

Very nicely done Alexandra! I wonder why the eye cannot pick this up visually. I wonder if the brain’s processing corrects for white balance. Or if it’s a contrast effect.

James

[EdAstle]

That's a result Alexandra! And top tip re levels. That inner glow really adds to already great images.

Photographers used to use a 50% grey card to set white balance. Digi cams now look for the "whitest" colours and readjust everything based on that, kinda. You've made a "solar grey" based on emittance rather than human perception. I think.

James - the brain is a fantastic "re-imaginer of truth". In a court of law it'd be called a liar.
I found a vid of my favourite optical illusion:
https://www.youtube.com/watch?v=GALLMJxLvgA

I've printed it out and cut the paper up and it's real.

"I couldn't believe my eyes!". Don't. Ever

[rigel123]

Those shots are pretty representative of what I saw visually with my 80mm and Lunt Wedge, other than the umbra being black visually.

And the color of the surface does match my description of vellum, not actually a color but the paper, vellum, is that color.

IMG_0060.jpeg (330.54 KiB) Viewed 3155 times

[rsfoto]

Hi,

I am somehow a bit sceptic about all these different colours of the sunspots. ¿?

If I look at the spectrum of the sun the highest luminance is in the green between 4900Å and 5400Å

So how is it possible that the sun spots are reddish brown or whatever colour?

This reminds me of the fact that the kids do paint the Sun yellow which is at the end of the green spectrum.

Sun-010SP-0013s-Stretch-graph.jpg (355.99 KiB) Viewed 3139 times

[DeepSolar64]

Rainer, all.
The best way to gauge the color of the Sun is probably visually. The problem here lies the extreme contrast between the photosphere, the sunspots and the black background outside the disk. It's a large dynamic range for the eye to adjust for. Another issue is one needs a filter that doesn't impart any color to the image. Most filters do. Even my Altair wedge is a warm white which is probably caused by the polarizing filter within it. Do they contain a neutral density filter as well, right? Very few filters show a truly natural unaffected spectrum. This may skew the subtle color of the photosphere as seen by eye or camera making the image look in most cases warmer than it actually is.

James

[DeepSolar64]

Everyone,
Today just before local noon I tried my Point Grey Flea color camera out on the white light photosphere using my little Orion 70mm refractor. The skies are mostly cloudy today with a lot of high and middle level clouds which makes observing and imaging difficult. The seeing also is not good. The temperature is 77F ( 25C ). I used the MacBook running ASTRO IIDC for image capture and the single captured image was edited in PS CC2020. The scope was mounted on the Losmandy AZ8.

Imaging!.JPG (10.69 KiB) Viewed 3068 times

I like the look of the disc in natural color, but I am not sure how " natural " it really is. Filters tend to add or subtract colors to an image and the image had a slight greenish tinge that is imparted to it by the glass over the aperture filter. I tried to reduce that a bit in PhotoShop. Otherwise I did not change the color at all. Still, the image is a bit warmer than it appears visually through an eyepiece. I don't think even most wedges have even color transmission through them since the remaining four percent of the light that passes the prism must be attenuated by a filter before it reaches the eye or camera. Filters are rarely color-neutral! And then there is the color calibration of your computer monitor to consider. Hmm...

Naturally this is true continuum. No eyepiece filter was used to narrow the bandpass. I can see some color differences in the spots and the photosphere. I may even see a tiny bit of a reddish tinge around the umbra of some of the larger spots. But is that color noise or is it truly there?

I'll do more experimenting in the future. I need good seeing and the C102 to do a close-up of a big spot region!!

James

20230823_15.43UTC_WLC_Color_Orion70_JP.jpg (3.57 MiB) Viewed 3069 times

[alessbonsai]

Rainer, all.
The best way to gauge the color of the Sun is probably visually. The problem here lies the extreme contrast between the photosphere, the sunspots and the black background outside the disk. It's a large dynamic range for the eye to adjust for. Another issue is one needs a filter that doesn't impart any color to the image. Most filters do. Even my Altair wedge is a warm white which is probably caused by the polarizing filter within it. Do they contain a neutral density filter as well, right? Very few filters show a truly natural unaffected spectrum. This may skew the subtle color of the photosphere as seen by eye or camera making the image look in most cases warmer than it actually is.

James

these are tests that I did between 13 and 14 approximately.
the colors are really hard to see and reproduce.
regarding the color temperature and the graph that rsfoto posted, i think yes, the sun emits all colors, but we have an outer layer of color that dims everything

2023-08-14-1123_5-L_lapl4_ap2235_IMPPG_PS.jpg (198.45 KiB) Viewed 3101 times

In my opinion, even if with different tones they are the most "real"

2023-08-13-1021_5-L_lapl4_ap2706IMPPG_PIX_PS.jpg (278.67 KiB) Viewed 3101 times

2023-08-11-0939_1-L_lapl4_ap19989IMPPG_PS.jpg (1.57 MiB) Viewed 3101 times

2023-08-11-0935_5-L_lapl4_ap1439_pix_PSOK.jpg (238.22 KiB) Viewed 3101 times

2023-08-11-0933_2-L_lapl4_ap1500imppgPS.jpg (121.3 KiB) Viewed 3101 times

2023-08-11-0933_2-L_lapl4_ap1500_PIX.jpg (174.56 KiB) Viewed 3101 times

[DeepSolar64]

Nice!
It's subtle but I can see a tiny bit of color shading around some of the umbrae in my images today that supports your findings. This seems to be surprisingly hard to see and image. The differences in color is small. I hope to do more work on this, time and weather permitting.

James

[DeepSolar64]

Today I took a few images in color of the solar disc using my Point Grey Flea color camera and Celestron 102mm Refractor. I think picking out color differences between the spots and the photosphere will be really difficult, at least with the equipment I have. To make it possible a highly color accurate filter is needed. One that does not add any color to the image. Even my wedge has a warm white tinge which adds a slight yellow warmness to the image. The camera seems to be more sensitive to this than the eye is with the eye seeing a less-yellow image. I think it would better be seen visually. It's worth a try.

Here are some images from today. Seeing wasn't good so sharp close ups were impossible.

James

20230831_15.18.10UTC_PGFleaColor_WL_C102_JP.jpg (1.53 MiB) Viewed 2998 times

20230831_15.25.42UTC_WL_PGFleaColor_C102_JP.jpg (1.1 MiB) Viewed 2998 times

20230831_15.26.27UTC_PGFleaColor_WL_C102_JP.jpg (1.13 MiB) Viewed 2998 times

[daniele_bonfiglio]

Dear Christian and all,
yesterday I finally had the opportunity to test myself against this amazing challenge you proposed.
I used a 127 mm Maksutov-Cassegrain telescope with a white light astrosolar filter in front, and of course a color sensor (a mirrorless camera Fujifilm X-T30).
During the post-processing with Photoshop, I applied a similar approach as for the mineral Moon images, namely starting from the stack with Autostakkert I sharpened the luminosity channel only and increased the saturation of the color channel. Regarding the colors I did a manual white balance by selecting as neutral color a region of the photosphere away from any sunspot. I did not apply any atmospheric dispersion correction.
To better appreciate the sunspot colors, I also made a version without sharpening of the luminosity channel, and also another one where instead of the luminosity channel taken from the stack I used a layer with uniform luminosity, so that the sunspots do not appear dark any longer.
Below are the results both cropped to the sunspot region and not. I think that the different colors of the umbra and penumbra regions with respect to the surrounding photosphere are evident, in particular in the image versions without sharpening and with uniform luminosity layer. In the non-cropped images, the change of colors towards the limb is also evident. I think this can be explained as related to the limb darkening, because the shallower layers of the sun sampled close to the limb are not only darker but also colder and therefore redder.
Thanks again Christian for this great challenge!
Best,
Daniele

No crop, with sharpening of the luminosity channel:

sunspot_colors_sharpened.jpg (1.91 MiB) Viewed 2731 times

.
No crop, without sharpening:

sunspot_colors_no_sharpening.jpg (1.37 MiB) Viewed 2731 times

.
No crop, with the luminosity channel replaced by a uniform luminosity value:

sunspot_colors_uniform_luminosity.jpg (513.9 KiB) Viewed 2731 times

.
Cropped, with sharpening of the luminosity channel:

sunspot_colors_crop_sharpened.jpg (306.83 KiB) Viewed 2731 times

.
Cropped, without sharpening:

sunspot_colors_crop_no_sharpening.jpg (224.59 KiB) Viewed 2731 times

.
Cropped, with the luminosity channel replaced by a uniform luminosity value:

sunspot_colors_crop_uniform_luminosity.jpg (70.2 KiB) Viewed 2731 times

[daniele_bonfiglio]

This is an addition to my previous post. I have further processed the last image of my previous post (with constant luminosity and thus only colour variations) by making a different white balance and an additional increase of saturation.
The result is below and shows that if the colour of the regions away from the sunspots is set to white, then the penumbra looks sort of light orange and the umbra is either orange or red depending on the size of the sunspots. Of course this is only qualitative but I think it gives the feeling of how the relative colours change in the different regions.
Best, Daniele

sunspot_colors_crop_uniform_luminosity-01.jpeg (94.86 KiB) Viewed 2722 times

sunspot_colors_comparison_full.gif (1.28 MiB) Viewed 2716 times

[christian viladrich]

Dear Christian and all,
Regarding the colors I did a manual white balance by selecting as neutral color a region of the photosphere away from any sunspot. I did not apply any atmospheric dispersion correction.
T
Daniele

Dear Daniel,

I think you've got some usefull data there.

Yet, a word of caution, for first image is not white balance. It has a huge blue outbalance. Just check the bottom left corner. It should be white while it is blue.

Can you try again ? You can use Astrosurface software to make the white balance based on the bottom left corner of the image.

It would be nice to share :
- first, the orignal data after correction of the white balance (i.e. no color saturation). Use the bottom/left corner for white balance
- then, the same with increased saturation of colors.

Both without sharpening / and then with sharpening.

Could you expand a little bit more on the "uniform luminosity layer". It sounds a bit like using a mask to reduce the strong difference of luminosity between the umbra and the photosphere ?

Did you start with RAW or jpg files ? 8 bits, 12 bits, or ? Number of frames used ?

[daniele_bonfiglio]

Dear Christian and all,
Regarding the colors I did a manual white balance by selecting as neutral color a region of the photosphere away from any sunspot. I did not apply any atmospheric dispersion correction.
T
Daniele

Dear Daniel,

I think you've got some usefull data there.

Yet, a word of caution, for first image is not white balance. It has a huge blue outbalance. Just check the bottom left corner. It should be white while it is blue.

Can you try again ? You can use Astrosurface software to make the white balance based on the bottom left corner of the image.

It would be nice to share :
- first, the orignal data after correction of the white balance (i.e. no color saturation). Use the bottom/left corner for white balance
- then, the same with increased saturation of colors.

Both without sharpening / and then with sharpening.

Could you expand a little bit more on the "uniform luminosity layer". It sounds a bit like using a mask to reduce the strong difference of luminosity between the umbra and the photosphere ?

Did you start with RAW or jpg files ? 8 bits, 12 bits, or ? Number of frames used ?

Dear Christian, you are right about white balance. Tomorrow I will try again and answer your questions. In the meantime, you can have a look at my second message (just before yours) where I have already corrected the white balance although in the cropped image only
Best, Daniele

[DeepSolar64]

I wish I could figure out how to image the sunspot colors with a color camera without having to tinker with the image in processing so much. Some tinkering to eliminate color being added by the filter and to make the sunspot colors more obvious will almost certainly be necessary but too much processing can add colors and artifacts that are not there at all.

If I can get the time and the seeing I am not done with this. I need a closer up view than my little Orion70 will do. I need to use the C102 with a barlow before the camera to isolate a spot. And that takes good seeing to pull off.

And then there is visual.

Nice results, Daniele.

James

[christian viladrich]

Hello James,

No needs to do anything complex. You just have to :
- take an image (= a video) with your color camera,
- process it as usual with AutoStakkert, nothing more,
- then, the only thing additionnal to do is to do the white balance by selecting a small area at the center of the Sun. You can use Astrosurface software to do this.

When done, with Photoshop (for example) display the histrogram for the ombra, and for the penombra. Just like this :
http://www.astrosurf.com/uploads/monthl ... fc5478.png

This will help to check if there is a dominant color or not in these different places. In the example given, there is indeed a dominant red color.

A possible additional step is to increase saturation, but of course this won't be realistic. This would be only to make more obvious the difference of colors.

Another way to do it is by visual observation. I could make out the slighlty brownish color of the solar limb this way. But I was not sure to see anything definite in the prenombra and I found the ombra completely black.

Large spots are probably preferable.

Tell us what you get.

[DeepSolar64]

Thanks for the advice Christian. I’ll see if I can find a copy of Astrosurface to do the white balance.

I too have noticed a slightly brownish color to the limb darkening as well but have never noticed any color visually in the spots. But I have never really looked closely for it either. I am usually looking for detail.

James

[daniele_bonfiglio]

Dear Daniel,

I think you've got some usefull data there.

Yet, a word of caution, for first image is not white balance. It has a huge blue outbalance. Just check the bottom left corner. It should be white while it is blue.

Can you try again ? You can use Astrosurface software to make the white balance based on the bottom left corner of the image.

It would be nice to share :
- first, the orignal data after correction of the white balance (i.e. no color saturation). Use the bottom/left corner for white balance
- then, the same with increased saturation of colors.

Both without sharpening / and then with sharpening.

Could you expand a little bit more on the "uniform luminosity layer". It sounds a bit like using a mask to reduce the strong difference of luminosity between the umbra and the photosphere ?

Did you start with RAW or jpg files ? 8 bits, 12 bits, or ? Number of frames used ?

Dear Christian,
thank you for your interest in my experiment.
I tried again with white balance using the bottom left corner as you suggested. It is much better now I think. Let me share what you asked for:

1) the original data after correction of white balance without any color saturation and without sharpening:

sunspot_colors_bis_1.jpg (396.29 KiB) Viewed 2615 times

2) like 1 but with increased saturation of colors:

sunspot_colors_bis_2.jpg (1.13 MiB) Viewed 2615 times

3) the original data after correction of white balance without any color saturation (as in 1) but with sharpening and level adjustments:

sunspot_colors_bis_3.jpg (1.21 MiB) Viewed 2615 times

4) like 3 but with increased saturation of colors:

sunspot_colors_bis_4.jpg (1.58 MiB) Viewed 2615 times

Also, let me show the histogram in selected areas for image 3 (only sharpening but not increase of color saturation):

Surrounding photosphere:

histogram_photosphere.png (584.75 KiB) Viewed 2615 times

Penumbra:

histogram_penumbra.png (585.03 KiB) Viewed 2615 times

Umbra:

histogram_umbra.png (585.23 KiB) Viewed 2615 times

Animated gif:

sunspot_colors_bis_animation.gif (537.16 KiB) Viewed 2604 times

Now about the uniform luminosity layer. What I do is just to use the color channel from image 2 but as the luminosity channel to use a uniform shade of gray. So the only differences in the resulting image are from the color levels, not from the luminosity levels. This is the result:

5) like 2 but with uniform luminosity:

sunspot_colors_bis_5.jpg (425.87 KiB) Viewed 2615 times

I have put my working file in google drive so that you can play with it by opening in photoshop. By switching on and off alternatively the first two layers, and by switching on and off the smart filters associated with each layer, you can obtain all the different image versions 1-5 posted here. This is the link: https://drive.google.com/file/d/1qrdP2y ... sp=sharing

Finally, let me answer your questions:
- my mirrorless camera can shot in 16 bits RAW, and 8 bits jpg with either "fine" or "normal" quality. I usually shot jpg with fine quality (ie the highest available) which makes me possible to shot continuously at 10 fps. Of course when shooting jpg you have to set the white balance and other parameters. I use the standard "film simulation" so no HDR or things like this, and I make sure to use any white balance settings apart from auto WB to avoid the white balance to change between shots. Typically I use "daylight" white balance which is more or less OK and then I adjust WB during post-processing.
- this was the stack of 16 best frames out of 500 taken in a time window of 50 seconds. Maybe I could have stacked more frames to reduce the color noise, but I think its already fine. I could have even made two stacks one with less frames stacked to get best details for the luminance channel, and then another stack with more frames to be used for the chrominance channel.

Let me know if you all have any comments or suggestions!

Best, Daniele

PS thank you again for the challenge, I have learnt a lot and had a lot of fun!

[christian viladrich]

Beautifully done Danielle !!

The succession of histograms show clearly the increasing brownish color of the penombra and ombra.

The version with increased saturation makes these differences visible to the eye.

The trick with "uniform luminosity layer" is quite effective. It shows us what color would have the ombra, if it were as bright as the surrounding photosphere.

Well done !

Next question ... can we see this visually ?
A long time ago, famous solar observer Father Secchi reported seeing the redish color of the umbra with his refractor (Vatican observatory, I don't remember the apreture of his scope, maybe 200 mm ?).
On my side, I was not able to see any color at the eyepiece. Maybe we need a very large spot, and find some way of blocking the light of neighboring photosphere ?

[Rusted]

I haven't ploughed through the whole thread. So forgive me if this has been suggested before:

Use a reflecting telescope, of sufficiently long focus, to produce a prime focus image of ONLY one [large] spot's umbra/penumbra.
Without any surrounding solar surface material, to upset a camera's colour balance, only the solar protective film could apply false colour.
What colour does the camera see? Could the Mk1, human eye, be able to ascertain colour in the resulting image?
A high, local solar altitude observing site and time would avoid most atmospheric refraction/diffraction effects.

[DeepSolar64]

I agree. A high resolution closeup of a spot should be the best way to see color in it for it should minimize contrast from the surrounding photosphere. A long focus reflector should work great though I don't see why a long focus well corrected refractor wouldn't work equally as well.

Ditto on a high altitude site as well. To lessen atmospheric distortions and to improve seeing.

James

[daniele_bonfiglio]

I agree that a long focal length scope that would produce a high resolution closeup of a single spot should be a possible way to see the sunspot colors. Other things that come to my mind are making a sort of cardboard mask to put on the eyepiece to mask the neighboring photosphere, and/or using some sheet of transparent material to smooth out the image and reduce the contrast between photosphere and sunspot.

I have a couple of additional questions to Christian and all of you:
1) using the information from the histograms in my last post, it would be possible to get an estimate of the actual temperature of different regions? Of course we need at least a mean of calibration. Suppose we assume that we know the photosphere temperature and we use it as calibration, I wonder if form the histograms we can get the temperature of the penumbra and umbra regions.
2) I also wonder if a similar challenge could be applied to solar granules as well! We can expect that the borders of the granules (with colder gas going down) would be redder than the interior of the granules (with hotter gas going up). However, to show this at least photographically we need a proper resolution to resolve the granules, and a sensor much more sensitive to colors because of the much smaller color differences across the granules than between sunspots and the photosphere... Maybe after all instead of trying to use the color information from the sensor, it would be better to use the luminosity information and derive the temperature from the Stefan-Boltzmann law... (this also applies to the sunspots)

Best, Daniele