Siriusly twinkling

By Phil Plait | January 27, 2012 9:19 am

If you live nearly anywhere on Earth — those of you north of 73° you’re out of luck, but I’m guessing there aren’t many of you! — and look to the southeast shortly after sunset, you’ll see the figure of Orion. Follow the three belt stars to the east, and you’ll see a bright star: Sirius, the brightest star in the night sky. If it’s near the horizon, you may see it twinkling madly: flickering, dancing, perhaps even changing color.

This gave astronomer David Lynch an idea: take a time exposure of Sirius with a camera and telephoto, and purposely wiggle the mount. He tried it on January 4, 2012, and the result he got is actually quite lovely:

Isn’t that cool? As the vibrating camera caused the star to trail around, the changing colors got recorded along the track. The changing brightness of Sirius can be seen as well, as parts of the loop-de-loop fade and intensify.

The reason stars twinkle is because of our atmosphere: little blobs of air are constantly in motion. These air parcels act like lenses, and as light passes through them, the path of the ray gets bent a little bit. That’s what causes the dancing motion, the actual twinkling. Different colors get bent by different amounts (which is why prisms break up white light into separate colors).

While it’s beautiful to our eyes, twinkling is a major pain to astronomers. It blurs our images! That’s why we launch telescopes into space, or design fancy optics for ground-based telescopes to remove it. Twinkling is free, but correcting it sure ain’t.

Lynch has several websites loaded with interesting pictures he’s taken of nature, including Thule Scientific, Color and Light in Nature, and San Andreas Fault.

Image credit: David Lynch (used by permssion). Tip o’ the Snell’s Law to Earth Science Picture of the Day.


MORE ABOUT: Sirius, twinkle
ADVERTISEMENT

Comments (47)

Links to this Post

  1. Siriusly twinkling | Bad Astronomy | My Blog | January 27, 2012
  2. Friday links | February 3, 2012
  3. Sirius and Long Exposure Photography! « Dr. Erika Grundstrom | February 6, 2012
  1. Pete Jackson

    What a great idea! Really gives a great demonstration of the brightness and color changes.

    And it gives a new meaning to the term “Bad Astronomy”

  2. thetentman

    I first noticed this in a very remote spot in Maine. I noticed a bright star low on the eastern horizon that seemes to be changing color. Flashes of, white, red, blue and yellow that almost looked like a bright flickering set of LEDs. At first I thought it might be a tower of some sort but as I was in the Hundred Mile Wilderness and there are no structures near. Then I thought a satellite? Nah, who would put a light on a satellite? I noticed it moved with the plane of the ecliptic and so was forced to conclude that it was a star. I made my wife look. She saw it too. When we got home (NJ) I called the Planetary Society and they were no help at all. Then I called or emailed the Museum of Natural History in NY and Joe Rao told me about diffraction (? I think that’s what he called it). Very cool.

  3. Minos

    Sirius has been changing color pretty spectacularly in the early evening lately. It’s really cool to see it in long exposure!

  4. Ben

    Very nice, just goes to show how bad the atmospheric effects really are for an astronomer!

    FYI, the San Andreas Fault link is missing a period :)

  5. Dwight Beery

    I believe the word Joe Rao used for the primary cause of twinkling of bright stars seen most easily near the horizon is “refraction” or bending of light Earth’s atmosphere.

  6. Chris

    Wait I thought there were no green stars. Yes I realize this is an optical effect, but I’ve spent plenty of time looking up and never did I see it twinkle green. Was there some enhancement or are my eyes too slow?

  7. Fabio Miguez

    @Chris, just guessing here, but it could be that to see Sirius (or any other star) twinkle into the green, it must be very, very close to the horizon, going through tons (literally) of air, and most people don’t have that kind of line of sight.

  8. Pete Jackson

    @4 Chris: There are no green stars because stars have an approximately black-body spectrum which never looks green to the eye no matter what the effective temperature of the star.

    However, all visible stars have some green light contained in the total spectrum, and it will be seen if a prism or, in this case the atmosphere, sends only the green part of the spectrum your way, even if just momentarily.

  9. Cool info. It appeals to me as both a nerd and a photographer.

    In this article, your link to San Andreas Fault is missing the period following the “www.” Hence, the link doesn’t work.

  10. BOB

    For a moment I thought the image was taken by the director David Lynch. He’s into all kinds of other stuff, why not astronomy too?

  11. Aw crap! I was gonna make a crack about green stars and you guys are already all over it!
    [/pout]
    Never has crappy seeing looked so good!

  12. Wow. That’s very nice. I kind of want to paint that.

  13. I’ve always loved the rich, pure colors that you can see with a twinkling star, colors that for the most part don’t appear in the night sky. I’m wondering, is there another way to do this? Maybe with a clock drive set to quickly move the telescope? Of course, the path described by a star moving across the telescope’s field would be much shorter than the path captured here. Beautifully done!

  14. @2 TheTentman: If it was on the plane of the ecliptic, that’d point to it being a planet. Possibly Venus? If it was early in the evening (or morning), that’s where I’d put my money, since you said it was low on the horizon. Venus can scintillate like mad when it’s bright. I’ve mistaken Venus for the landing lights of an airplane before :)

  15. Which brings me to the question of why most other twinkling stars don’t appear to change color. Is the diffraction effect always present but simply more noticeable to the naked eye when the source (star, planet, whatever) is extremely bright?

  16. @10 Harold: Heh, I was having the same thought. If there were a way to make the camera/scope move in a spiral, you could capture a much longer period in one shot.

  17. thetentman

    It moved with all the other stars in the sky. Perhaps I phrased it wrong. Mr. Rao was even able to tell me what star I was seeing once I gave him the time and date.

  18. Tehanu

    I love watching Sirius dance every evening — I get home about 7:45 p.m. and right now it is fairly high in the eastern sky. Since I live in L.A. I don’t get to see a lot of stars but Sirius of course stands vividly out. Thanks for this, it’s wonderful.

  19. carmen

    Okay, I realise this might come as a shallow comment, as it’s based strictly on aesthetics and ignoring the sirius science-y stuff, but this reminds me of the ancient Ottoman sultans’ official seal, the Tughra, a beautifully crafted piece of calligraphy. http://en.wikipedia.org/wiki/Tughra

  20. amphiox

    Joseph @16;

    It most likely relates to brightness. IIRC, only the brightest stars are bright enough to stimulate our color sensitive retinal cone cells. The rest only stimulate our night vision adapted rods, which are monochrone.

  21. Aldebaraan

    Beautiful, I love it…and want to try it out for myself.

    One question, though. What are the small white dots scattered around the image? There is a bright one in the upper right, and 3 or 4 smaller ones around the image. They look large enough to be more than a single pixel large, so I wouldn’t think it’d be a CCD artifact (but I’m not an expert there). Ideas?

  22. Chris S

    @14 Harold, @17 Joseph G;

    I had the same thought, although I wondered if the camera could oscillated vertically and slowly turned at the same time to trace out a sinusoidal pattern (like a pure audio tone on an oscilloscope).

    I was thinking that you could then vary the frequency of the vertical oscillation to see if there were patterns in the changing of the colour. A repeated pattern of a particular colour on a regular interval would stand out easily in this way. Think of it as “physical visualization”.

    I sometimes use rapid vibration of my eyes to see otherwise invisible patterns. For example, some “purple” Christmas light strings are red/blue LEDs in parallel/same direction, while others are parallel opposite direction. By vibrating my eyes, I can tell whether the red/blue LEDs are flashing (at 60Hz) in phase or out of phase. This imaging of Sirius strikes me as a similar, camera-based, technique.

  23. Infinite123Lifer

    Peak interest.

    I spent all last night periodically stalking Sirius. The twinkle action was something fierce. There was no comparison for lively action twinkle elsewheres in the sky and it made me wonder if Sirius had a secret. I was squinting my eyes at it. Orion, the Moon and I think Venus were all running from Sirius last night. Last night & the night before were pretty amazing here by the end of the night with everything piling up on their way to the horizon together.

    Here Sirius seems to twinkle no matter where it is in the sky or dare I say which day?. Which, since the air don’t change around here much would make sense, but Iam going to pay particular attention to degree of twinkle according to its position all relative to my vantage point of course. Now all I need is my Bat-twinkle detector.

    I heard in some background noise on a tv about ancient people calling one star a Devil Star for its tendency to “blink”, haven’t researched this one yet but it made me wonder “could Sirius be the infamous (to me @ least) Devil Star?”

    On a less probably researchable notion: I wonder what Sirius looked like before present day air and light pollution.

  24. @21 Carmen: Okay, I realise this might come as a shallow comment, as it’s based strictly on aesthetics and ignoring the sirius science-y stuff, but this reminds me of the ancient Ottoman sultans’ official seal, the Tughra, a beautifully crafted piece of calligraphy. [link baleeted]

    Nothing shallow about that. Beautiful!

  25. @23 Chris: I was thinking that you could then vary the frequency of the vertical oscillation to see if there were patterns in the changing of the colour. A repeated pattern of a particular colour on a regular interval would stand out easily in this way. Think of it as “physical visualization”.
    Cool! I know it’s possible, I just don’t know enough about optics to know how…

    @24Infinite: I heard in some background noise on a tv about ancient people calling one star a Devil Star for its tendency to “blink”, haven’t researched this one yet but it made me wonder “could Sirius be the infamous (to me @ least) Devil Star?”

    Well, Algol (Beta Persei) is known in many cultures as the “demon star,” and in fact the name Algol comes from the Arabic “al-ghūl,” or literally, “the ghoul” (the English word “ghoul” deriving from the Arabic). It’s an eclipsing binary, so its brightness varies: every two days and 20 hours, there’s a 10-hour long eclipse of Beta Persei A by the dimmer Beta Persei B. At least one star in the system also experiences extensive flaring from the interaction of the two stars’ magnetic fields, though I doubt this is measurable with backyard equipment.

  26. Sam H

    This is a very interesting concept, with a beautiful result :). The above comments tend to agree with my view, although this image has many different meanings to many different people…

    For me though, it can only have one 😉

    http://www.youtube.com/watch?v=OmyPSCLCPss

  27. One of the first things I look for in the night sky is the triangle of Rigel, Sirius and Canopus. Three of the brightest stars in the sky (hence among the first to come out), and depending where they are in the sky, the angle at Sirius can look close to a right angle (actually it’s somewhat wider). Facing in the other direction is the triangle of Rigel, Sirius and Procyon (which is closer to an actual right angle, but Procyon’s not as bright as the others).

    The line running approximately through Canopus, Sirius and Procyon, I once nicknamed the False Ecliptic. It’s not straight, but it’s straight enough that if you were in a strange city and tried to locate the ecliptic by naiively lining up the brightest objects in the sky, you could easily think that was it.

    (You can tell I’m in the Southern Hemisphere.)

  28. STARLIGHT IS COHERENT

    Various things confused me about starlight, twinkle, etc., but my misconceptions evaporated when I really grasped the idea that starlight is coherent. Not monochromatic of course, yet starlight far more coherent than laser light. The plane-wave wavefront is all in phase, same as with parallel laser light. That’s why stellar interferometry works. Even Betelgeuse, the least coherent example, has a coherence length measured in meters. Stars are coherent point-sources, so their light produces several phenomena seen with laser illumination. Twinkle, for example.

    STAR TWINKLE IS DRIFTING “LASER SPECKLE” AND DIFFRACTION FRINGES

    Twinkle is not from refraction exactly, instead it’s interference patterns. If starlight was bright enough to cast shadows, we’d see that it creates a color-confetti version of laser speckle on all surfaces it illuminates. Refraction comes in when the moving air with nonuniform density causes starlight to slightly bend, and so the laser-speckle patterns are drifting along as the air-patterns drift. Cross your eyes at a twinkling star, and you’ll notice that the two images don’t twinkle in sync. The patterns of starlight dancing on your face has structures smaller than the distance between your eyes. Planets? That’s always explained wrong. Visible planets aren’t point-sources, aren’t coherent. Their light has too short coherence length, so, like a light bulb, they don’t produce the speckle and fringes which we see as twinkle. Go light up Pluto with a few H-bombs, then you’d probably see some twinkle.

    The above photo (*very* cool btw) is a time-sampling of the moving patterns intercepted by the camera lens.

    And so, what would a nearby nova look like? Besides being a blinding star in the sky, it might be bright enough to visibly light up surfaces at night. And in that case, we’d finally be able to see the “twinkle patterns” directly. How will they appear? They move with wind, so probably they’d be motion-blurred to the eye, and only be slow enough on fairly still nights. I suspect they’d resemble the sunlight patterns on the bottom of a swimming pool. But with lots of diffraction fringes added. And with the entire pattern drifting along with air motion.

    Pretty cool, eh?

    Maybe someone somewhere with a high speed image-intensifier camera has recorded the moving starlight patterns on a white surface at night?

    PS
    During an eclipse, in the few seconds before totality, the apparent source has the shape of a slit: wide in one direction and point-source in the other. Rather than causing “laser speckle,” this should cause parallel diffraction fringes all over the ground, and should be bright enough to be seen by eye. Jail bars moving on the ground, moving at the speed of the wind. (Probably they’re always there just before totality, but motion-blurred unless wind conditions are just right.)

  29. Great idea, great result, great write-up. Cheers! :-)

    @25. Infinite123Lifer : “On a less probably researchable notion: I wonder what Sirius looked like before present day air and light pollution.”

    It might be a very long trip to find out but there are still areas of desert and ocean with no light pollution and minimal air pollution. Antartica, the Great Sandy Desert here in Oz, the Sahara and so on. From there even from just slightly light & air polluted regions the stars are spectacular indeed. :-)

    (Not that I’ve ever been to any of thsoe mentioned there save the very last one. Best skies I’ve ever seen were from near a hamlet along the Murray river with the Flinder’s ranges and Eyre Peninsula skies as runners up.)

    @27. Joseph G : Good answer there with Algol – I definitly agree there. Isaac Asimov used that “demon star” idea in one of his short mystery stories too from memory. :-)

  30. @30. bill beaty :

    And so, what would a nearby nova look like? Besides being a blinding star in the sky, it might be bright enough to visibly light up surfaces at night. And in that case, we’d finally be able to see the “twinkle patterns” directly.

    It would depend on just how close that nova (or supernova or hypernova!) was – and we don’t want one right next door – like thirty or so light years distant – as that’d bring Death from the Skies!

    Stellar expert James B. Kaler notes that if Betelgeux were to go supernovae tonight :

    .. it would become as bright as a gibbous Moon, would cast strong shadows on the ground, and would be seen easily in full daylight.

    With Betelgeux being about 570 light-years away approximately. Source – Betelguese page, Kaler’s Stars page – is linked to my name here.

    @7. Chris : Wait I thought there were no green stars.

    Actually there’s at least one – Beta Librae or Zubeneschamali is often thought to be green and sometimes looks that way to me. 8)

    Click on the link then go to Kaler’s ‘Stars’ page from there and find Zubeneschamali on his list and you’ll see him refer to that too. There’s also Antares b and a few other green companion stars which may or may not be just down to contrast effects. The BA has discussed the topic of green stars here a few times too.

    @3. thetentman : “Then I thought a satellite? Nah, who would put a light on a satellite?”

    The International Space Station crew. It helps when they put the ship’s cat out so it can return okay! 😉

    Seriously, I’d suggest looking again the following few nights in around the same place and /or using some good planetarium software or a planisphere. You should be able to work out what it is from those. If it moves relative to other stars and twinkles less than them especially when higher up its probably a planet. If not, probably a bright star. Hope that helps. :-)

  31. For Zubeneschamali – See :

    http://stars.astro.illinois.edu/sow/zubenes.html

    See also :

    http://blogs.discovermagazine.com/badastronomy/2008/08/28/followup-green-objects-in-space/

    &

    http://blogs.discovermagazine.com/badastronomy/2011/11/22/the-green-ghost-of-a-distant-dead-star/

    for a few discussions on this green stars issue.

    Plus there’s the ‘WISE finds the coolest stars. Literally.’ item which is linked to my name for this comment although the green star there is very much false colour.

  32. CR

    This is kind of off topic, but I never miss an opportunity to bring up the tv show Space: 1999 on this blog… the Algol/blinking demon reference reminded me that the 1970’s comic books based upon S99 had an issue that featured the wayward moon travelling near a binary star system, a red giant and a blue dwarf. The local inhabitants of a planet orbiting the stars were affected by the presence of the blue dwarf when it wasn’t eclipsed by the red giant… their ancestors had made statues of their gods doing battle, the two stars represented as the eyes of a winking demon. One of the Alphan characters made reference to Earth’s old cultural references to Algol. The issue title was even ‘Demon Star.’
    Imagine that… a comic based upon a tv show, and it had a little educational/historical material in its script. Yeah!

    As for the blinking of Sirius, when I was not quite a teenager, I mistook the blinking as lights on an aircraft, but realised that even a helicopter wouldn’t hold so steady in the sky for so long and figured it was a star. I figured that it wasn’t an alien UFO, either, for pretty much the same reason. (Well, that and I already was starting to doubt the whole ‘alien spacecraft’ notion of UFOs, no matter how much I wanted there to be alien life out there!) Shortly thereafter, I started learning the constellations, and brighter stars, solving the ‘mystery’ I had witnessed.

    As for the srtistic way of showing off Sirius’ colorful activity in our night sky: STUNNING!

  33. PJF

    @30 Bill Beaty: STARLIGHT IS COHERENT
    Various things confused me about starlight, twinkle, etc., but my misconceptions evaporated when I really grasped the idea that starlight is coherent. Not monochromatic of course, yet starlight far more coherent than laser light. The plane-wave wavefront is all in phase, same as with parallel laser light. That’s why stellar interferometry works. Even Betelgeuse, the least coherent example, has a coherence length measured in meters. Stars are coherent point-sources, so their light produces several phenomena seen with laser illumination. Twinkle, for example.
    STAR TWINKLE IS DRIFTING “LASER SPECKLE” AND DIFFRACTION FRINGES…

    …And so, what would a nearby nova look like? Besides being a blinding star in the sky, it might be bright enough to visibly light up surfaces at night. And in that case, we’d finally be able to see the “twinkle patterns” directly. How will they appear? They move with wind, so probably they’d be motion-blurred to the eye, and only be slow enough on fairly still nights. I suspect they’d resemble the sunlight patterns on the bottom of a swimming pool. But with lots of diffraction fringes added. And with the entire pattern drifting along with air motion.
    Pretty cool, eh?

    Whoah. You seriously just blew my mind!!!

    Maybe someone somewhere with a high speed image-intensifier camera has recorded the moving starlight patterns on a white surface at night?

    Are there instruments sensitive enough to do that?? I’d think the big problem would be that the patterns from all the stars would overlap and look more like TV static, if that. Perhaps you could use a box with a small hole, like a pinhole camera, so that you’d only be photographing the light from a few stars (presumably one bright one)? Of course with a small aperture like that, the amount of light you’re getting would go from miniscule to incomprehensibly tiny…

  34. @32 MTU: It helps when they put the ship’s cat out so it can return okay!

    Heh, thanks to that I did a youtube search for zero-G cats. Apparently I’m not the only one who thought of it. It’s also apparent that it’s an extremely bad idea. No gravity means nothing keeping those claws away from your face :)

  35. @30 Bill Beaty: STARLIGHT IS COHERENT
    Various things confused me about starlight, twinkle, etc., but my misconceptions evaporated when I really grasped the idea that starlight is coherent. Not monochromatic of course, yet starlight far more coherent than laser light. The plane-wave wavefront is all in phase, same as with parallel laser light. That’s why stellar interferometry works. Even Betelgeuse, the least coherent example, has a coherence length measured in meters. Stars are coherent point-sources, so their light produces several phenomena seen with laser illumination. Twinkle, for example.
    STAR TWINKLE IS DRIFTING “LASER SPECKLE” AND DIFFRACTION FRINGES…

    …And so, what would a nearby nova look like? Besides being a blinding star in the sky, it might be bright enough to visibly light up surfaces at night. And in that case, we’d finally be able to see the “twinkle patterns” directly. How will they appear? They move with wind, so probably they’d be motion-blurred to the eye, and only be slow enough on fairly still nights. I suspect they’d resemble the sunlight patterns on the bottom of a swimming pool. But with lots of diffraction fringes added. And with the entire pattern drifting along with air motion.
    Pretty cool, eh?

    Whoah. You seriously just blew my mind!!!

    Maybe someone somewhere with a high speed image-intensifier camera has recorded the moving starlight patterns on a white surface at night?

    Are there instruments sensitive enough to do that?? I’d think the big problem would be that the patterns from all the stars would overlap and look more like TV static, if that. Perhaps you could use a box with a small hole, like a pinhole camera, so that you’d only be photographing the light from a few stars (presumably one bright one)? Of course with a small aperture like that, the amount of light you’re getting would go from miniscule to incomprehensibly tiny…

  36. Infinite123Lifer

    Interesting stuff there. Thanks for clearing that up Joseph G. I kept procrastinating on that one. Note, demon star not devil star. Got it. Algol.

    “It might be a very long trip to find out but there are still areas of desert and ocean with no light pollution and minimal air pollution. Antartica, the Great Sandy Desert here in Oz, the Sahara and so on. From there even from just slightly light & air polluted regions the stars are spectacular indeed. :-)

    Doh, I just went to a Q&A at the university with photographer, science writer, oceanographer and much more Chris Linder. He has spent a bit of time down in Antarctica photographing penguin colonies as he explained and though tangential to the topic of penguins I still missed some great questions. “What did the sky look like?” and “What difficulties did you ever have in trying to view(if ever) the nighttime sky in such an environment?” Evidently clear and beautiful would be the obvious answers. Still, would have liked to have gotten a response from someone who had just recently been there.

    Maybe next time.

    I had a question about skies, Mauna Loa Observatory, minimal air pollution, mixing of air on the local level versus the global level, air pollution now and way before, but I think I will hold off on those, I am having a tough time wording my question and when that happens it is usually because I have not researched my topic thoroughly enough :)

  37. WTF happened with my posts?

  38. @Infinite123: I did some more Googling, and it looks like some people do refer to it as the devil star. Demon, devil, potato, potahto… :)

  39. Infinite123Lifer

    potato, potahto. got it :)

    Now to locate said ghoul-ey light-tress. :)

  40. I’ve been under the apparently mistaken impression that the reason stars twinkle and planets seldom do is because the amplitude of the light waves from stars exceeds the arc or area of sky that the object fills. Stars that look bigger aren’t necessarily bigger or closer. Proxima Centauri certainly doesn’t look bigger than Sirius.

    If twinkle is caused solely by refraction then shouldn’t artificial satellites and the planets also twinkle? I understand that the inconsistencies in the atmosphere (density, temperature, wind and particulate matter) can and will cause varying refraction of light and sound waves. But the observable difference between star twinkle and planetary steadiness (comparatively), both refracting through the same atmosphere, suggests that refraction is not the (sole) cause of twinkle. Neither does the fact that planetary light is indirect/reflected light from our Sun doesn’t seem to satisfactorily explain the difference. This leaves me feeling a bit in the dark… word play, bazinga.

  41. Joseph G

    @42 T. Ray: I thought they did? I’ve seen Venus twinkle a heck of a lot. Or maybe we have different definitions of “twinkle”? :)

  42. Maybe bright planets like venus aren’t quite point-sources and the twinkle has different qualities because of it? Hrm.

  43. Dan C

    Wouldn’t the speed the camera is moving also explain some of the variation in light intensity in the photo. If you notice, the tips of the ‘leaves’ tend to be brighter than the middle of them, because the camera is moving slower at the end of its arc.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

ADVERTISEMENT

See More

ADVERTISEMENT
Collapse bottom bar
+