20Jun2017 Spectroheliograms: Hydrogen
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
20Jun2017 Spectroheliograms: Hydrogen
Greetings Everyone.
Slow to process the results from my 20June imaging session.
Here are results of hydrogen alpha and beta.
An animation giving a spectral sweep through the H alpha line.
Numbers are offsets from line centre in milli-Angstroms.
Cheers.
Peter.
Slow to process the results from my 20June imaging session.
Here are results of hydrogen alpha and beta.
An animation giving a spectral sweep through the H alpha line.
Numbers are offsets from line centre in milli-Angstroms.
Cheers.
Peter.
Last edited by p_zetner on Fri Jul 07, 2017 2:23 pm, edited 1 time in total.
- marktownley
- Librarian
- Posts: 42638
- Joined: Tue Oct 18, 2011 5:27 pm
- Location: Brierley Hills, UK
- Has thanked: 20940 times
- Been thanked: 10587 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
These are excellent Peter! I really do like the sweep!
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!
-
- Way More Fun to Share It!!
- Posts: 2040
- Joined: Sat Nov 24, 2012 2:33 pm
- Has thanked: 57 times
- Been thanked: 1208 times
- Montana
- Librarian
- Posts: 34764
- Joined: Mon Oct 17, 2011 5:25 pm
- Location: Cheshire, UK
- Has thanked: 18101 times
- Been thanked: 8935 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
WOW!! these are the best images I have ever seen and the animation really puts it in context
Alexandra
Alexandra
- Carbon60
- Way More Fun to Share It!!
- Posts: 14335
- Joined: Wed Mar 07, 2012 12:33 pm
- Location: Lancashire, UK
- Has thanked: 8588 times
- Been thanked: 8324 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
This is really excellent work, Peter. Well done!
Cool Dopplergram. It would be really interesting to see how the pattern changes over time. Maybe use 'difference' imaging to show which regions are changing more rapidly than others. Fascinating stuff.
Stu.
Cool Dopplergram. It would be really interesting to see how the pattern changes over time. Maybe use 'difference' imaging to show which regions are changing more rapidly than others. Fascinating stuff.
Stu.
H-alpha, WL and Ca II K imaging kit for various image scales.
Fluxgate Magnetometers (1s and 150s Cadence).
Radio meteor detector.
More images at http://www.flickr.com/photos/solarcarbon60/
Fluxgate Magnetometers (1s and 150s Cadence).
Radio meteor detector.
More images at http://www.flickr.com/photos/solarcarbon60/
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Alexandra and Mark: Many thanks for your kind comments!
highfnum: I'm slowly getting better at improving image quality both with hardware and software.
Stu: Many thanks for your comments and your excellent suggestion about difference imaging! In fact, most of my interest in this and knowledge of it comes from a famous paper by Leighton, Noyes and Simon ("Velocity Fields in the Solar Atmosphere I. Preliminary Report") who performed observations exactly along the lines you're suggesting. The dopplergram maps out velocity fields while dopplergram difference images would map out acceleration fields. Here's a link to the paper:
http://adsabs.harvard.edu/abs/1962ApJ...135..474L
(Click on " Full Refereed Journal Article (PDF/Postscript)".)
My problem with implementing this is that I have a lot of difficulty with combining spectroheliograms obtained on successive sweeps (something I would have to do to produce a difference dopplergram). Successive sweeps never give exactly the same shape of the solar disk, for example. I'm beginning to suspect that the (real) camera frame rate is not exactly constant. I'm using a PGR camera, PGR acquisition software and have tried both "buffered" and "streaming" modes of video acquisition. At some point, I will switch to Firecapture or Genika to see if this problem can be overcome.
Maybe you'd be interested in some earlier analysis I carried out of the (hydrogen alpha) chromospheric velocity field in an active region (when there were active regions!). I was able to capture some "usual" H-alpha images and some H-alpha spectroheliograms in the same observing session. I overlayed the (colourized) dopplergram on the higher resolution "usual" image to try and match flow velocities with observed features.
Here are the results.
The first montage of images shows various images obtained during the session in comparison to the NSO-GONG full disk. The big active region was the "monster" AR12529. The second panel of images shows how I colourized the dopplergram and overlayed it onto the original "hi-res" image and its inverted counterpart. One feature that's striking is the presence of high velocity plasma (both outflows, in blue, and inflows, in orange) near the outer boundary of the spot penumbra. This is in line with observations first made by Leighton et al in their paper. Also, if you look in the bottom, left corner of the overlayed images (SW corner in correct orientation) you can see the footpoint of a filament which exhibits a spectacular "jumble" of outflow and inflow!
Hope you find this interesting!
Many thanks, again, for your comments.
Peter.
highfnum: I'm slowly getting better at improving image quality both with hardware and software.
Stu: Many thanks for your comments and your excellent suggestion about difference imaging! In fact, most of my interest in this and knowledge of it comes from a famous paper by Leighton, Noyes and Simon ("Velocity Fields in the Solar Atmosphere I. Preliminary Report") who performed observations exactly along the lines you're suggesting. The dopplergram maps out velocity fields while dopplergram difference images would map out acceleration fields. Here's a link to the paper:
http://adsabs.harvard.edu/abs/1962ApJ...135..474L
(Click on " Full Refereed Journal Article (PDF/Postscript)".)
My problem with implementing this is that I have a lot of difficulty with combining spectroheliograms obtained on successive sweeps (something I would have to do to produce a difference dopplergram). Successive sweeps never give exactly the same shape of the solar disk, for example. I'm beginning to suspect that the (real) camera frame rate is not exactly constant. I'm using a PGR camera, PGR acquisition software and have tried both "buffered" and "streaming" modes of video acquisition. At some point, I will switch to Firecapture or Genika to see if this problem can be overcome.
Maybe you'd be interested in some earlier analysis I carried out of the (hydrogen alpha) chromospheric velocity field in an active region (when there were active regions!). I was able to capture some "usual" H-alpha images and some H-alpha spectroheliograms in the same observing session. I overlayed the (colourized) dopplergram on the higher resolution "usual" image to try and match flow velocities with observed features.
Here are the results.
The first montage of images shows various images obtained during the session in comparison to the NSO-GONG full disk. The big active region was the "monster" AR12529. The second panel of images shows how I colourized the dopplergram and overlayed it onto the original "hi-res" image and its inverted counterpart. One feature that's striking is the presence of high velocity plasma (both outflows, in blue, and inflows, in orange) near the outer boundary of the spot penumbra. This is in line with observations first made by Leighton et al in their paper. Also, if you look in the bottom, left corner of the overlayed images (SW corner in correct orientation) you can see the footpoint of a filament which exhibits a spectacular "jumble" of outflow and inflow!
Hope you find this interesting!
Many thanks, again, for your comments.
Peter.
- Montana
- Librarian
- Posts: 34764
- Joined: Mon Oct 17, 2011 5:25 pm
- Location: Cheshire, UK
- Has thanked: 18101 times
- Been thanked: 8935 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Wow! (d) is a very good image indeed, it is really interesting to see one side there is inflow and the other is outflow. I wonder if this is the same in every spot or different if it is reversed (which side) in opposite polarity spots. Great work!! please do more
Alexandra
Alexandra
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Alexandra.
Thanks for commenting!
Images (d) and (e) give the same information (the colour scheme is the same). I just found it easier, sometimes, to view the colour distribution against a light background as opposed to a dark (the footpoint of the filament, for example).
As far as the flow velocities near a spot, I can't say it any better than the original observers of the phenomenon!
From Leighton et al (link above):
"Near sunspots, we find the region inside the outer boundary of the penumbra to be
quiescent, just as for the lower levels seen in the Ca 6103 and Na 5896 plates. In Ha,
however, the area immediately outside this boundary is one of enhanced motions, which
consists mainly of an inward flow of streams of material, accelerating as it approaches
the penumbra and halting abruptly at the boundary. Occasionally, one sees outward moving
material, generally coming from this same boundary, traveling at relatively high
velocity; the outward progress of this ejected matter can be followed on plates taken a
few minutes apart. We are evidently seeing here, on the disk, motions which appear as
quiescent and eruptive prominences on the limb. In a few cases, an eruptive feature
will be light along one edge and dark along the other, as if the structure were in rapid
axial rotation."
All the best.
Peter.
Thanks for commenting!
Images (d) and (e) give the same information (the colour scheme is the same). I just found it easier, sometimes, to view the colour distribution against a light background as opposed to a dark (the footpoint of the filament, for example).
As far as the flow velocities near a spot, I can't say it any better than the original observers of the phenomenon!
From Leighton et al (link above):
"Near sunspots, we find the region inside the outer boundary of the penumbra to be
quiescent, just as for the lower levels seen in the Ca 6103 and Na 5896 plates. In Ha,
however, the area immediately outside this boundary is one of enhanced motions, which
consists mainly of an inward flow of streams of material, accelerating as it approaches
the penumbra and halting abruptly at the boundary. Occasionally, one sees outward moving
material, generally coming from this same boundary, traveling at relatively high
velocity; the outward progress of this ejected matter can be followed on plates taken a
few minutes apart. We are evidently seeing here, on the disk, motions which appear as
quiescent and eruptive prominences on the limb. In a few cases, an eruptive feature
will be light along one edge and dark along the other, as if the structure were in rapid
axial rotation."
All the best.
Peter.
- marktownley
- Librarian
- Posts: 42638
- Joined: Tue Oct 18, 2011 5:27 pm
- Location: Brierley Hills, UK
- Has thanked: 20940 times
- Been thanked: 10587 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
All interesting stuff Peter! You are doing some excellent science there!
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!
- Carbon60
- Way More Fun to Share It!!
- Posts: 14335
- Joined: Wed Mar 07, 2012 12:33 pm
- Location: Lancashire, UK
- Has thanked: 8588 times
- Been thanked: 8324 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Slow to respond here, Peter, due to illness, but this is excellent work which provides further insight into the dynamics of our Sun. I've often noticed the apparent inflow of matter towards the penumbra on some of the spectacular animations shared on this forum, which, with outflow displayed at the same time, was a little confusing. I thought the phenomenon might have been the result of narrowband imaging artefacts. Your images clearly show, however, that both flows can and do occur simultaneously in these turbulent and magnificent 3D structures. Nicely done.
Stu.
Stu.
H-alpha, WL and Ca II K imaging kit for various image scales.
Fluxgate Magnetometers (1s and 150s Cadence).
Radio meteor detector.
More images at http://www.flickr.com/photos/solarcarbon60/
Fluxgate Magnetometers (1s and 150s Cadence).
Radio meteor detector.
More images at http://www.flickr.com/photos/solarcarbon60/
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Thanks Mark and Stu.
Stu: Thanks for sharing your observations. I'm going to have to go back and view some of those animations a bit more critically! Hope you're feeling better.
Cheers.
Peter.
Stu: Thanks for sharing your observations. I'm going to have to go back and view some of those animations a bit more critically! Hope you're feeling better.
Cheers.
Peter.
- Phil Rousselle
- Oh, I get it now!
- Posts: 26
- Joined: Thu Feb 27, 2014 5:39 pm
- Location: France
- Been thanked: 7 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Peter, Very good work !
I would like to know a little bit more. may I ask few questions :
- What is the original size of the disk (height in pixels) ?
- Do you work with capture in 8, 10, 12 bits ?
- What is your method to reduce "transversalium" lines (it looks like what i use) ?
- Do you create your own processing software ?
- For Dopplergram, what is the spread between Ha+ and Ha- ?
Ok I stop here
Regards
Phil
I would like to know a little bit more. may I ask few questions :
- What is the original size of the disk (height in pixels) ?
- Do you work with capture in 8, 10, 12 bits ?
- What is your method to reduce "transversalium" lines (it looks like what i use) ?
- Do you create your own processing software ?
- For Dopplergram, what is the spread between Ha+ and Ha- ?
Ok I stop here
Regards
Phil
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Phil.
Many thanks for your visit and comments. As far as your questions go:
- What is the original size of the disk (height in pixels) ?
The original disk size in vertical (non-dispersive) direction is ~ 2050 px.
- Do you work with capture in 8, 10, 12 bits ?
At the moment, I only use 8-bit depth but will extend this in future imaging sessions. I am using a PGR Grasshopper 6.0Mpx (ICX694) camera with a 200px x 2736px region of interest selected, running at 50 frames per second.
- What is your method to reduce "transversalium" lines (it looks like what i use) ?
I use a number of methods, sometimes in combination (blended with Photoshop)!
(1) "Wah" method. (accomplished with ImageJ)
a) crop central (feature-free, if possible) portion of original spectroheliogram (uncorrected for aspect ratio, dimensions are 15100px x 2736px).
b) shrink cropped image in non-dispersion direction (horizontal in my case) to vertical column of 1px (horizontal width) to 2736px (vertical
width) dimension.
c) reset aspect ratio to expand to original dimensions (15100px x 2736px) to produce flat.
d) divide original spectroheliogram by flat.
(2) "Rousselle" method (partially).
a) same as steps a) and b) above.
b) fit 1px column to polynomial (4th to 6th order) and calculate residuals.
c) convert polynomial fit to 15100px x 2736px dimensions. This is flat to correct intensity variation in vertical (non-dispersion) direction.
d) convert fit residuals to 15100px x 2736px dimensions. Subtract this from original spectroheliogram to eliminate transversalium.
e) sometimes I repeat steps a), b), c) to correct intensity variations in the horizontal (dispersion) direction.
Unlike you, I don't usually flat-field the dispersion direction (step e)). I find it's the non-dispersion direction that is most affected by vignetting,
scatter, etc.
(3) ImageJ FFT line suppression.
If I'm lucky and the slit remains relatively clean, I can often get away with the "suppress lines" (horizontal) feature in the ImageJ FFT bandpass
filter. This is less successful with bright features present as often the case in CaK images.
- Do you create your own processing software ?
To process the assembled spectroheliogram, I use a combination of ImageJ, Photoshop and, when necessary for fitting, Origin. Straightening and selecting the spectral line from the original avi video is accomplished with Virtual Dub.
- For Dopplergram, what is the spread between Ha+ and Ha- ?
The dopplergram that I presented is an average of 5 dopplergrams taken at +/- 0.28, 0.37, 0.46, 0.55 and 0.65 Angstroms. I couldn't find any significant difference between any of these individual dopplergrams, hence, the average. This is not strictly true for the active region on the limb. There seemed to be some small variation with detuning in the dopplergram. I'll explore this more when I can study active regions at higher spatial resolution.
Cheers.
Peter.
Many thanks for your visit and comments. As far as your questions go:
- What is the original size of the disk (height in pixels) ?
The original disk size in vertical (non-dispersive) direction is ~ 2050 px.
- Do you work with capture in 8, 10, 12 bits ?
At the moment, I only use 8-bit depth but will extend this in future imaging sessions. I am using a PGR Grasshopper 6.0Mpx (ICX694) camera with a 200px x 2736px region of interest selected, running at 50 frames per second.
- What is your method to reduce "transversalium" lines (it looks like what i use) ?
I use a number of methods, sometimes in combination (blended with Photoshop)!
(1) "Wah" method. (accomplished with ImageJ)
a) crop central (feature-free, if possible) portion of original spectroheliogram (uncorrected for aspect ratio, dimensions are 15100px x 2736px).
b) shrink cropped image in non-dispersion direction (horizontal in my case) to vertical column of 1px (horizontal width) to 2736px (vertical
width) dimension.
c) reset aspect ratio to expand to original dimensions (15100px x 2736px) to produce flat.
d) divide original spectroheliogram by flat.
(2) "Rousselle" method (partially).
a) same as steps a) and b) above.
b) fit 1px column to polynomial (4th to 6th order) and calculate residuals.
c) convert polynomial fit to 15100px x 2736px dimensions. This is flat to correct intensity variation in vertical (non-dispersion) direction.
d) convert fit residuals to 15100px x 2736px dimensions. Subtract this from original spectroheliogram to eliminate transversalium.
e) sometimes I repeat steps a), b), c) to correct intensity variations in the horizontal (dispersion) direction.
Unlike you, I don't usually flat-field the dispersion direction (step e)). I find it's the non-dispersion direction that is most affected by vignetting,
scatter, etc.
(3) ImageJ FFT line suppression.
If I'm lucky and the slit remains relatively clean, I can often get away with the "suppress lines" (horizontal) feature in the ImageJ FFT bandpass
filter. This is less successful with bright features present as often the case in CaK images.
- Do you create your own processing software ?
To process the assembled spectroheliogram, I use a combination of ImageJ, Photoshop and, when necessary for fitting, Origin. Straightening and selecting the spectral line from the original avi video is accomplished with Virtual Dub.
- For Dopplergram, what is the spread between Ha+ and Ha- ?
The dopplergram that I presented is an average of 5 dopplergrams taken at +/- 0.28, 0.37, 0.46, 0.55 and 0.65 Angstroms. I couldn't find any significant difference between any of these individual dopplergrams, hence, the average. This is not strictly true for the active region on the limb. There seemed to be some small variation with detuning in the dopplergram. I'll explore this more when I can study active regions at higher spatial resolution.
Cheers.
Peter.
-
- Way More Fun to Share It!!
- Posts: 2040
- Joined: Sat Nov 24, 2012 2:33 pm
- Has thanked: 57 times
- Been thanked: 1208 times
- Phil Rousselle
- Oh, I get it now!
- Posts: 26
- Joined: Thu Feb 27, 2014 5:39 pm
- Location: France
- Been thanked: 7 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Peter,
thank you very much for your extensive reply.
Your "raw image" is very large (15000 x 2700 pixels) and video must be more than 8Go !. My processing software cannot "eat" such a big image. Maybe after a binning it could be working.
Can't you see Doppler effect induced by the rotational motion ? and Zeeman effect on sunspots ?
Congratulations again
Phil
thank you very much for your extensive reply.
Your "raw image" is very large (15000 x 2700 pixels) and video must be more than 8Go !. My processing software cannot "eat" such a big image. Maybe after a binning it could be working.
Can't you see Doppler effect induced by the rotational motion ? and Zeeman effect on sunspots ?
Congratulations again
Phil
-
- Way More Fun to Share It!!
- Posts: 2161
- Joined: Sun Jun 14, 2015 4:46 pm
- Location: France
- Has thanked: 1 time
- Been thanked: 2725 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Peter,
This is beautifull !
Can you post or send me some of the individual frames from -0.5 A to 0.5 A ?
What is the bandpass of your spectroheliograph ?
Thanks
Christian
This is beautifull !
Can you post or send me some of the individual frames from -0.5 A to 0.5 A ?
What is the bandpass of your spectroheliograph ?
Thanks
Christian
Christian Viladrich
Co-author of "Planetary Astronomy"
http://planetary-astronomy.com/
Editor of "Solar Astronomy"
http://www.astronomiesolaire.com/
Co-author of "Planetary Astronomy"
http://planetary-astronomy.com/
Editor of "Solar Astronomy"
http://www.astronomiesolaire.com/
- Merlin66
- Librarian
- Posts: 3972
- Joined: Tue Dec 13, 2011 1:23 pm
- Location: Junortoun, Australia
- Has thanked: 174 times
- Been thanked: 619 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Most if not all the current “cutting edge” SHG results and instrumentation is well covered in my latest book “Imaging Sunlight”
I thank all the amateurs around the world who contributed information on their instruments and details of their processing techniques.
Recommended to all solar observers.
I thank all the amateurs around the world who contributed information on their instruments and details of their processing techniques.
Recommended to all solar observers.
"Astronomical Spectroscopy - The Final Frontier" - to boldly go where few amateurs have gone before
https://groups.io/g/astronomicalspectroscopy
http://astronomicalspectroscopy.com
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer
https://groups.io/g/astronomicalspectroscopy
http://astronomicalspectroscopy.com
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hello Phil and Christian.
Thank you both for your kind comments. There is more information about my SHG in a somewhat recent article I wrote for the Spectroscopy section of the "Amateur Astronomy Association Germany (VdS)".
The article is in Spektrum No.52 - 2017 (http://spektroskopie.fg-vds.de/index_e.htm).
The direct link to the Journal issue: http://spektroskopie.fg-vds.de/pdf/Spektrum52.pdf
Phil:
The VirtualDub program seems to handle large video files well. Once the video is loaded into VDub, it is frame-by-frame cropped to isolate the spectral line and the set of individual cropped strips is exported to a folder. I then use ImageJ to import the strips into a stack and the "Montage" function assembles the strips without a problem. My operating system (Windows 7) is 64 bit and ImageJ is the compatible 64 bit version. I seem to recall "out of memory" problems with the 32 bit version of ImageJ.
I have studied the magnetic field near a sunspot and measured solar differential rotation spectroscopically. Here are links to the studies:
Sunspot Magnetic Field: https://www.cloudynights.com/topic/5723 ... tic-field/
Solar Differential Rotation: https://www.cloudynights.com/topic/5745 ... eliograph/
Christian:
Glad to see you here on the Spectroscopy thread! The bandpass of the SHG used for the images presented here is about 0.22 Angstroms. For the Sunspot Magnetic Field analysis linked above, this was reduced to 0.085 Angstroms with a narrower entrance slit to the spectrometer. Below, I'm posting a "montage" of every second frame of the animation above. Let me know if you'd like me to send you individual images.
Best Regards.
Peter.
Thank you both for your kind comments. There is more information about my SHG in a somewhat recent article I wrote for the Spectroscopy section of the "Amateur Astronomy Association Germany (VdS)".
The article is in Spektrum No.52 - 2017 (http://spektroskopie.fg-vds.de/index_e.htm).
The direct link to the Journal issue: http://spektroskopie.fg-vds.de/pdf/Spektrum52.pdf
Phil:
The VirtualDub program seems to handle large video files well. Once the video is loaded into VDub, it is frame-by-frame cropped to isolate the spectral line and the set of individual cropped strips is exported to a folder. I then use ImageJ to import the strips into a stack and the "Montage" function assembles the strips without a problem. My operating system (Windows 7) is 64 bit and ImageJ is the compatible 64 bit version. I seem to recall "out of memory" problems with the 32 bit version of ImageJ.
I have studied the magnetic field near a sunspot and measured solar differential rotation spectroscopically. Here are links to the studies:
Sunspot Magnetic Field: https://www.cloudynights.com/topic/5723 ... tic-field/
Solar Differential Rotation: https://www.cloudynights.com/topic/5745 ... eliograph/
Christian:
Glad to see you here on the Spectroscopy thread! The bandpass of the SHG used for the images presented here is about 0.22 Angstroms. For the Sunspot Magnetic Field analysis linked above, this was reduced to 0.085 Angstroms with a narrower entrance slit to the spectrometer. Below, I'm posting a "montage" of every second frame of the animation above. Let me know if you'd like me to send you individual images.
Best Regards.
Peter.
-
- Way More Fun to Share It!!
- Posts: 2161
- Joined: Sun Jun 14, 2015 4:46 pm
- Location: France
- Has thanked: 1 time
- Been thanked: 2725 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Thanks a lot Peter !
As we go from -0.55A to +0.55A, could we say we have about 0.09 A shift between two successive images ?
It seems to me that the three images close to the centre line are very similar. In other words, beeing +or - 0.09A from the center line gives the same result ?
Do you have an idea of the transmission profile of the spectroheliograph ? Fabry-Perot etalon have a Lorentzian profile, what about spectro helio ? Are they more selective for the same FWHM ?
Best regards
Christian
As we go from -0.55A to +0.55A, could we say we have about 0.09 A shift between two successive images ?
It seems to me that the three images close to the centre line are very similar. In other words, beeing +or - 0.09A from the center line gives the same result ?
Do you have an idea of the transmission profile of the spectroheliograph ? Fabry-Perot etalon have a Lorentzian profile, what about spectro helio ? Are they more selective for the same FWHM ?
Best regards
Christian
Christian Viladrich
Co-author of "Planetary Astronomy"
http://planetary-astronomy.com/
Editor of "Solar Astronomy"
http://www.astronomiesolaire.com/
Co-author of "Planetary Astronomy"
http://planetary-astronomy.com/
Editor of "Solar Astronomy"
http://www.astronomiesolaire.com/
- Merlin66
- Librarian
- Posts: 3972
- Joined: Tue Dec 13, 2011 1:23 pm
- Location: Junortoun, Australia
- Has thanked: 174 times
- Been thanked: 619 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Christian,
I'm sure Peter will answer...
Based on my SHG trials the ability to extract a nominated wavelength (based on the pixel column of the acquired image) determines both the central wavelength (after calibration) and the resolution element (ie 0.09A etc)
As we use a narrow slit in the SHG the resulting spectral line profile is Gaussian. (We/I use a neon or Ar reference lamp to establish the FWHM and resulting resolution of the spectrograph in normal operation)
Hope that helps.
(resolution v's dispersion and FWHM reference measurements are discussed in my book "Imaging Sunlight" - Sect 6.2.1, Sect 6.7.7 and Sect 7.4. An interesting comparison is shown on p199 - a pair of CaK spectroheliograms based on 1 pixel and ten pixel selection v's a similar image from the CaK PST.)
I'm sure Peter will answer...
Based on my SHG trials the ability to extract a nominated wavelength (based on the pixel column of the acquired image) determines both the central wavelength (after calibration) and the resolution element (ie 0.09A etc)
As we use a narrow slit in the SHG the resulting spectral line profile is Gaussian. (We/I use a neon or Ar reference lamp to establish the FWHM and resulting resolution of the spectrograph in normal operation)
Hope that helps.
(resolution v's dispersion and FWHM reference measurements are discussed in my book "Imaging Sunlight" - Sect 6.2.1, Sect 6.7.7 and Sect 7.4. An interesting comparison is shown on p199 - a pair of CaK spectroheliograms based on 1 pixel and ten pixel selection v's a similar image from the CaK PST.)
"Astronomical Spectroscopy - The Final Frontier" - to boldly go where few amateurs have gone before
https://groups.io/g/astronomicalspectroscopy
http://astronomicalspectroscopy.com
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer
https://groups.io/g/astronomicalspectroscopy
http://astronomicalspectroscopy.com
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer
- Spectral Joe
- Ohhhhhh My!
- Posts: 120
- Joined: Wed Mar 14, 2012 4:34 am
- Location: Livermore, California
Re: 20Jun2017 Spectroheliograms: Hydrogen
Here is a spectral scan of a spectroheliograph with a 0.3 Angstrom bandpass, the red curve is the data and the blue curve is a Gaussian fit. This is the instrumental line width, matching the entrance slit and pixel size to the Airy disc size. Narrowing the entrance slit further will only reduce the throughput, the width at the exit plane will stay the same.
Observing the Sun with complex optical systems since 1966, and still haven't burned, melted or damaged anything.
Not blind yet, either!
Light pollution? I only observe the Sun, magnitude -26.74. Pollute that!
Not blind yet, either!
Light pollution? I only observe the Sun, magnitude -26.74. Pollute that!
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Christian.
In answer to your questions:
>>>As we go from -0.55A to +0.55A, could we say we have about 0.09 A shift between two successive images ?
Yes, the change in wavelength between frames is 0.092A.
>>>It seems to me that the three images close to the centre line are very similar. In other words, beeing +or - 0.09A from the center line gives the same result ?
Yes, in my experience this is generally true. An exception might be if there is a feature with large radial velocity in which case the appearance of the feature might look significantly different on the long wavelength vs short wavelength side of line centre.
>>>Do you have an idea of the transmission profile of the spectroheliograph ? Fabry-Perot etalon have a Lorentzian profile, what about spectro helio ?
The spectroheliograph lineshape is that of a grating spectrometer with slit-width limited resolution. The mathematical shape of such a line is determined by the entrance slit width, the associated angular spread arising from the diffraction pattern of this slit introduced by the collimator optics, the convolution of this spread function with the ideal grating function (sin^2(N*delta)/sin^2(delta)) and another convolution with the diffraction pattern from the camera optics! A wide slit produces a corresponding wide spectral line and, as the slit width is decreased, the spectral linewidth also shrinks until diffraction effects begin to dominate.
There is a discussion of the lineshape function in "Spectrophysics" by Thorne, Litzen and Johansson (Springer 1999) Chapter 12. They show that the lineshape evolves from a flat-topped function at large slit width to a smoother, peaked function for smaller slits. They describe an optimum slit size condition in which the image slit width (equal to physical slit width for collimator focal length equal to camera focal length) matches the diffraction limited width of the camera optics. These authors show a number of calculated lineshapes for various slit widths and, for narrower widths, the function certainly appears to resemble Gaussian (their figure 12.2).
Eversberg and Vollman ("Spectroscopic Instrumentation" Springer 2015) demonstrate that the sequence of convolutions required to mathematically determine the lineshape function can ultimately reduce to a Gaussian lineshape (although this argument depends on some fairly significant approximations).
Being somewhat unhappy about the detail required to mathematically determine a lineshape and not quite convinced about the approximations introduced by Eversberg and Vollman, I decided to perform a simple measurement of lineshapes for various slit widths (notwithstanding the statement of Ken Harrison and the nice picture of Spectral Joe).
The setup is pictured. The SHG telescope was removed and a Hg discharge tube was placed in front of the slit holder. I imaged the Hg yellow doublet (5769.598A, 5790.663A), captured in a single frame of the video camera, to provide a convenient wavelength calibration and do a lineshape study on the stronger component (5790.663A) of the doublet.
Here are the results for slit widths of 100 microns (line fwhm ~ 0.7 A), 25 microns (line fwhm = 0.21 A) and 10 microns (line fwhm = 0.14 A).
Black squares are measured data and red curves represent curves of best fit.
You can see the flat-topped lineshape at 100 microns (obviously not a Gaussian profile!) reduces to smoother functions at 10 and 25 microns which seem to be well-described by Gaussians. I was planning to include a 50 micron slit measurement but discovered that the bayonet holder was badly machined and didn't fit into my instrument. In any case, I would say that my observations are completely in line with the theory of Thorne et al. Also, it appears that, for a small enough slit width, the lineshape is Gaussian.
Best Regards.
Peter.
In answer to your questions:
>>>As we go from -0.55A to +0.55A, could we say we have about 0.09 A shift between two successive images ?
Yes, the change in wavelength between frames is 0.092A.
>>>It seems to me that the three images close to the centre line are very similar. In other words, beeing +or - 0.09A from the center line gives the same result ?
Yes, in my experience this is generally true. An exception might be if there is a feature with large radial velocity in which case the appearance of the feature might look significantly different on the long wavelength vs short wavelength side of line centre.
>>>Do you have an idea of the transmission profile of the spectroheliograph ? Fabry-Perot etalon have a Lorentzian profile, what about spectro helio ?
The spectroheliograph lineshape is that of a grating spectrometer with slit-width limited resolution. The mathematical shape of such a line is determined by the entrance slit width, the associated angular spread arising from the diffraction pattern of this slit introduced by the collimator optics, the convolution of this spread function with the ideal grating function (sin^2(N*delta)/sin^2(delta)) and another convolution with the diffraction pattern from the camera optics! A wide slit produces a corresponding wide spectral line and, as the slit width is decreased, the spectral linewidth also shrinks until diffraction effects begin to dominate.
There is a discussion of the lineshape function in "Spectrophysics" by Thorne, Litzen and Johansson (Springer 1999) Chapter 12. They show that the lineshape evolves from a flat-topped function at large slit width to a smoother, peaked function for smaller slits. They describe an optimum slit size condition in which the image slit width (equal to physical slit width for collimator focal length equal to camera focal length) matches the diffraction limited width of the camera optics. These authors show a number of calculated lineshapes for various slit widths and, for narrower widths, the function certainly appears to resemble Gaussian (their figure 12.2).
Eversberg and Vollman ("Spectroscopic Instrumentation" Springer 2015) demonstrate that the sequence of convolutions required to mathematically determine the lineshape function can ultimately reduce to a Gaussian lineshape (although this argument depends on some fairly significant approximations).
Being somewhat unhappy about the detail required to mathematically determine a lineshape and not quite convinced about the approximations introduced by Eversberg and Vollman, I decided to perform a simple measurement of lineshapes for various slit widths (notwithstanding the statement of Ken Harrison and the nice picture of Spectral Joe).
The setup is pictured. The SHG telescope was removed and a Hg discharge tube was placed in front of the slit holder. I imaged the Hg yellow doublet (5769.598A, 5790.663A), captured in a single frame of the video camera, to provide a convenient wavelength calibration and do a lineshape study on the stronger component (5790.663A) of the doublet.
Here are the results for slit widths of 100 microns (line fwhm ~ 0.7 A), 25 microns (line fwhm = 0.21 A) and 10 microns (line fwhm = 0.14 A).
Black squares are measured data and red curves represent curves of best fit.
You can see the flat-topped lineshape at 100 microns (obviously not a Gaussian profile!) reduces to smoother functions at 10 and 25 microns which seem to be well-described by Gaussians. I was planning to include a 50 micron slit measurement but discovered that the bayonet holder was badly machined and didn't fit into my instrument. In any case, I would say that my observations are completely in line with the theory of Thorne et al. Also, it appears that, for a small enough slit width, the lineshape is Gaussian.
Best Regards.
Peter.
-
- Way More Fun to Share It!!
- Posts: 2161
- Joined: Sun Jun 14, 2015 4:46 pm
- Location: France
- Has thanked: 1 time
- Been thanked: 2725 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Thanks a lot Peter, Joe, and Ken,
Indeed, I was wondering how the transmission curve of the SH compares to the Lorentzian transmission of the F-P etalon.
I've got the measurements of the transmission of two different 0.3A FWHM etalons. I'll post them this week-end.
Indeed, I was wondering how the transmission curve of the SH compares to the Lorentzian transmission of the F-P etalon.
I've got the measurements of the transmission of two different 0.3A FWHM etalons. I'll post them this week-end.
Christian Viladrich
Co-author of "Planetary Astronomy"
http://planetary-astronomy.com/
Editor of "Solar Astronomy"
http://www.astronomiesolaire.com/
Co-author of "Planetary Astronomy"
http://planetary-astronomy.com/
Editor of "Solar Astronomy"
http://www.astronomiesolaire.com/
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Hi Christian.
You got me interested in the SHG lineshape problem and I will do some more modelling calculations shortly!
It seems pretty clear, though, that, under optimum conditions, the SHG lineshape will be Gaussian.
For your interest in comparing the selectivity of a Gaussian vs Lorentzian line, I'm attaching a quick comparison calculation.
This picture shows Lorentzian and Gaussian functions of equal fwhm, normalized to give the same peak height of unity. The horizontal axis gives the offset from line centre (at zero) in units of fwhm. You can see that the Gaussian clearly drops to zero faster than the Lorentzian and, hence, might be considered to be more selective (in the sense of eliminating photospheric continuum).
It's possible that larger slit widths might introduce even steeper line "shoulders", dropping more rapidly than the simple Gaussian. This might give even more selectivity. I'm looking into this.
Best Regards.
Peter.
You got me interested in the SHG lineshape problem and I will do some more modelling calculations shortly!
It seems pretty clear, though, that, under optimum conditions, the SHG lineshape will be Gaussian.
For your interest in comparing the selectivity of a Gaussian vs Lorentzian line, I'm attaching a quick comparison calculation.
This picture shows Lorentzian and Gaussian functions of equal fwhm, normalized to give the same peak height of unity. The horizontal axis gives the offset from line centre (at zero) in units of fwhm. You can see that the Gaussian clearly drops to zero faster than the Lorentzian and, hence, might be considered to be more selective (in the sense of eliminating photospheric continuum).
It's possible that larger slit widths might introduce even steeper line "shoulders", dropping more rapidly than the simple Gaussian. This might give even more selectivity. I'm looking into this.
Best Regards.
Peter.
- Merlin66
- Librarian
- Posts: 3972
- Joined: Tue Dec 13, 2011 1:23 pm
- Location: Junortoun, Australia
- Has thanked: 174 times
- Been thanked: 619 times
- Contact:
Re: 20Jun2017 Spectroheliograms: Hydrogen
Thanks for that Peter!
I think what you are saying is appropriate to slit resolution (standard spectrograph) and is also reflected in the reference emission lines used for resolution measurements. Most SHG work with minimal gap sizes - lot's of light to play with!
I also think that the bandwidth selected (# columns selected) and the subsequent processing (combining/ stacking) influence the outcome in images produced by the SHG.
I think what you are saying is appropriate to slit resolution (standard spectrograph) and is also reflected in the reference emission lines used for resolution measurements. Most SHG work with minimal gap sizes - lot's of light to play with!
I also think that the bandwidth selected (# columns selected) and the subsequent processing (combining/ stacking) influence the outcome in images produced by the SHG.
"Astronomical Spectroscopy - The Final Frontier" - to boldly go where few amateurs have gone before
https://groups.io/g/astronomicalspectroscopy
http://astronomicalspectroscopy.com
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer
https://groups.io/g/astronomicalspectroscopy
http://astronomicalspectroscopy.com
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer
- p_zetner
- Way More Fun to Share It!!
- Posts: 1644
- Joined: Sun Mar 18, 2012 4:59 pm
- Location: Winnipeg, Manitoba, Canada
- Has thanked: 166 times
- Been thanked: 719 times
Re: 20Jun2017 Spectroheliograms: Hydrogen
Some last words on the SHG lineshape function.
Here is a figure showing model calculations of the line profile of a grating spectrometer for various slit widths.
These lineshape functions were calculated by performing a convolution of a box function representing the slit transmission (no diffraction) with a Gaussian function (4 pixels fwhm) representing the diffraction limited performance of the spectrometer. The choice of a Gaussian function was made based on the arguments in section 2.6.5 of Eversberg and Vollman ("Spectroscopic Instrumentation" Springer 2015).
The Gaussian function was chosen to have a fwhm of 4 pixels, consistent with measurements made using my SHG. For any given instrument, the fwhm of this limiting Gaussian is dependent on the optical speed of the collimator and camera lens systems as well as the illuminated area and groove density of the grating.
The figure is consistent with figure 12.2 in "Spectrophysics" by Thorne, Litzen and Johansson (Springer 1999). The flat topped lineshape moves to a smoother function with decreasing slit width but its fwhm is ultimately limited by diffraction, as we'd expect, and decreasing slit width below the diffraction limit will only reduce instrumental throughput without affecting spectral resolution.
The next figure shows a Gaussian fit performed on the "10 px image slit" profile in the previous figure.
You can see that a Gaussian fit to the real (modelled) lineshape function is not bad. The flat-topped feature of the real lineshape function is apparent but the fwhm and overall shape of the real function don't depart too severely from the Gaussian shape. The Gaussian shape is, thus, a pretty good approximation to the real lineshape for reasonably small slits.
Cheers.
Peter
Here is a figure showing model calculations of the line profile of a grating spectrometer for various slit widths.
These lineshape functions were calculated by performing a convolution of a box function representing the slit transmission (no diffraction) with a Gaussian function (4 pixels fwhm) representing the diffraction limited performance of the spectrometer. The choice of a Gaussian function was made based on the arguments in section 2.6.5 of Eversberg and Vollman ("Spectroscopic Instrumentation" Springer 2015).
The Gaussian function was chosen to have a fwhm of 4 pixels, consistent with measurements made using my SHG. For any given instrument, the fwhm of this limiting Gaussian is dependent on the optical speed of the collimator and camera lens systems as well as the illuminated area and groove density of the grating.
The figure is consistent with figure 12.2 in "Spectrophysics" by Thorne, Litzen and Johansson (Springer 1999). The flat topped lineshape moves to a smoother function with decreasing slit width but its fwhm is ultimately limited by diffraction, as we'd expect, and decreasing slit width below the diffraction limit will only reduce instrumental throughput without affecting spectral resolution.
The next figure shows a Gaussian fit performed on the "10 px image slit" profile in the previous figure.
You can see that a Gaussian fit to the real (modelled) lineshape function is not bad. The flat-topped feature of the real lineshape function is apparent but the fwhm and overall shape of the real function don't depart too severely from the Gaussian shape. The Gaussian shape is, thus, a pretty good approximation to the real lineshape for reasonably small slits.
Cheers.
Peter