A “light echo” from a star actually shows light moving

November 10, 2013 • 12:00 pm

What you’re seeing in the 50-second video below is not a star exploding, but the movement of light from a star as it bounces off dust and reaches us on Earth. That star emitted a burst of light in 2002, which, as it moved outward, illuminated parts of the dust cloud surrounding the star.  (No, the cloud isn’t moving.) As the light reaches parts of the dust cloud farther from the star, it gets reflected back to Earth in sequence. (By the way, if you want to spend an instructive 15 minutes, read the Wikipedia article on the speed of light, which is pretty good. Did you know that light moves through a diamond at one-third of the speed it moves in a vacuum?)

The video gives four years of time-lapse photos from the Hubble Space Telescope. Since light travels 186,000 miles per second (nearly 300,000 km/sec), you can see how vast space is compared to the speed of light.  The YouTube description is below.

The unusual variable star V838 Monocerotis (V838 Mon) continues to puzzle astronomers. This previously inconspicuous star underwent an outburst early in 2002, during which it temporarily increased in brightness to become 600,000 times more luminous than our Sun. Light from this sudden eruption is illuminating the interstellar dust surrounding the star, producing the most spectacular “light echo” in the history of astronomy.

As light from the eruption propagates outward into the dust, it is scattered by the dust and travels to the Earth. The scattered light has travelled an extra distance in comparison to light that reaches Earth directly from the stellar outburst. Such a light echo is the optical analogue of the sound echo produced when an Alpine yodel is reflected from the surrounding mountainsides.

The NASA/ESA Hubble Space Telescope has been observing the V838 Mon light echo since 2002. Each new observation of the light echo reveals a new and unique “thin-section” through the interstellar dust around the star. This video morphs images of the light echo from the Hubble taken at multiple times between 2002 and 2006. The numerous whorls and eddies in the interstellar dust are particularly noticeable. Possibly they have been produced by the effects of magnetic fields in the space between the stars.

credit: ESA/Hubble

h/t: Michael

70 thoughts on “A “light echo” from a star actually shows light moving

  1. I don’t understand this. If we’re seeing light from the outburst propagate across the dust cloud that light could not have reached Earth yet? How can we be seeing this at all?


    1. The light reached the dust cloud (comparatively) shortly after it left the star, and then it’s been bounced in our direction and is just arriving here now.

      Imagine you’re many miles out at sea away from a lighthouse, and you’re watching the beam sweep across a cloud near the light; that’s the sort of thing you’re seeing here.


        1. Yes, it does. However, I’ve heard them come up with bizarre answers to it involving the light source being manipulated or something (can’t remember – my take away was they even distorted that if IIRC).

          1. Ah yes, of course. The conclusion is so perfect too. Silly scientists. Why try to figure stuff out when you “weren’t there”. Just use your bible. Also, pray for airplanes to fly and illness to be cured. 😀 Awesome.

            That is why, ultimately, the only way to know about the past for certain is to have a reliable historic record written by an eyewitness. That is exactly what we have in the Bible.

    2. The star is 20,000 ly away so the light left the star 20,000 years ago. In other words, we’re just seeing what happend ~20,000 years ago.

      Kind of like if Betelgeuse goes nova – maybe it already has but it will take ~642 years for the light to reach us.

      1. I sort of worry about Betelgeuse. After the light from that one reaches us (and it might at any time), I think we are going to have some very bright days and bright nights for a while. Any info about how that might effect our weather?

        1. I seem to recall that there was fear about Betelgeuse sending out gamma rays our way that would pretty much sterilize the planet, however I think that has been put to rest. I also seem to recall that the brightness would be as bright as earth’s moon for some time but not as bright as the sun.

          1. “I seem to recall that there was fear about Betelgeuse sending out gamma rays our way”

            From what I understand, Betelgeuse is too far to hurt us when it goes, although it will probably be pretty bright. Seems like I remember hearing that a supernova would have to be within 20-50 light years to negatively affect us.

            Now, a gamma ray burst caused by a star collapsing into a black hole (which are literally cosmically rare) would sterilize us, even from hundreds or thousands of light years away, assuming one of its poles was pointed at us.

          2. I’m not aware of a determination of the orientation of the rotation axis of Betelgeuse compared to the plane of the sky (our POV). Can you cite a source (approximately) ; I can envisage a number of measurements that would potentially reveal this datum, but as I say, I haven’t actually noticed news of anyone being successful.
            (I see a report in “Universe Today” making the assertion that we’re not looking down Betelgeuse’s axis ; but no grounds to support that assertion are given.)

          3. “Because the spectra were obtained scanning across the stellar image, it is possible to measure a differential line shift across the disk of the star that can be interpreted as being due to rotation. If this interpretation is correct, we can determine the axis of rotation of Betelgeuse and estimate its rotational speed to be 1.2 × 10-8 rad s-1, corresponding to a rotation period of 17 yr. In addition, it is plausible that the 1995 March bright spot is congruent with the pole of the star, suggesting that star’s angle of inclination is ∼20° to the line of sight.

            Considering the small number of bright spots that are present at any one time on the surface of Betelgeuse, and the signature of the 1995 March spot in the Mg II resonance lines as observed with the GHRS, it appears that such spots are not the consequence of convective flows.”

            [ http://iopscience.iop.org/1538-3881/116/5/2501/fulltext/980184.text.html ]

          4. Sigh, I forgot to update. My excuses for the double ref.

            Oh well, at least I added some info.

        1. Galactic President Zaphod Beeblebrox would be rather annoyed! Given that the stars poles aren’t pointing at us it will be a great show for us mere hitch hikers. The neutrinos [& presumable the gravity waves too] would arrive 2-3 hours before the visible light thus we’d have time to point our optical telescopes & miss nothing:- Our Galaxy’s Next Supernova

          Speculation:- It would shine brighter than the full moon for a few weeks

          1. Great clip ~ those curtains are criminal though. Not one dud scene in that movie, helped no doubt by being pre-CGI when tight dialogue & plot ruled the process. CGI has ruined modern cinema for me.

    3. The huge burst in the apparent brightness of the star occurred 20,000 years ago [the star is 20,000 light years away] & the increased light only begun to arrive in the Hubble lens [in Earth orbit] 11 years ago. In the video we are watching 4 years of activity from 2002 to 2006 compressed into 50 seconds.

  2. This can really help put Deep Time into perspective. The stars look so immovable, so fixed, so perfect…and, yet, the galaxy (and the universe) is really a seething cauldron of activity. It is only because we are so small and, like mayflies, our lives are so short, that we do not see the roiling of the Cosmos.


    1. Deep Time? The star is barely into the double digits of Myr. It’s a stripling. Our last common ancestor with the gorillas could well have looked up into a sky without a Betelgeuse in it.
      Walk up a thousand metres of sandstone hillside in Scotland, each couple of millimetres having taken a few years to form, then look over the landscape beyond to see that this sandstone is deposited in the valleys of a landscape that predates it, and is overlain by another similar thickness body of sandstones, which themselves are overlain by kilometres of rock pushed into place on faults (while the whole lot was under 3-5km of confining pressures – we can do geothermobarometry on the fault plane’s melt rocks to determine that. And all of those overlying rocks have been carved up into the kilometre-scale relief that you’ve just climbed up. That’s “Deep Time.”
      I’ve got that point across to people on two two different occasions at two different locations (on the same fault plane ; 60-odd km apart). Only one of them was literally knocked off his feet – the desired response. Because the other had sat down through my exposition of the landscape’s history laid out before him.

      1. Well, sure, if you want to club them over the head with it, rather than start with baby steps.

        The Grand Canyon is another great place for the type of layered geology you describe, of course. Actually, if you ever make it to the American Southwest, I’d love to hang around near your elbow there….


        1. Pretty unlikely that I’d be in the American SW. For a start, it’s desert, and relatively hot. So you’d need to be paying me to go there.
          Yeah, it is clubbing them over the head. Since the people in question have asked me to teach them about geology … well, I feel no guilt over it.

          1. Well, the Canyon is desert, yes, but high desert. The rim is about 7,000 feet above sea level. Even in summer, the daytime high at the rim is generally in the mid-80s. At the river, the high can reach triple digits, but only during the hottest part of the day during the hottest part of the year — and, even then, it still cools off to the 70s overnight. The North Rim completely shuts down during the winter months due to snow, and they frequently have to close the South Rim as well. Right now, it’s 63°F and sunny at the Canyon.


  3. I see at least 6 faces in the cloud. Is that god and some angels? Is Chopra right about a living universe?

    /snarky sarcasm

  4. As noted by JAC the video is displaying the movement of light over a period of four years in only 50 seconds…

    Thus we are being shown the expansion of the hemispherical* “light front” sped up by a factor of approx 2,500,000 & yet at these astronomical scales light still appears to be a lazy slow coach

    * With the inside surface facing us. The lit up dust we observe here is nearly all behind the star, not in front

    1. “The lit up dust we observe here is nearly all behind the star, not in front”

      Okay, it’s all starting to make sense now. The light bounces off the dust behind the cloud and then shoots straight at us.

      But how do you know that the dust is behind the cloud?

      1. We know that the dust is behind the star because we see its light after the star’s flare. Dust in front of the star would have been lit by light that was travelling toward us anyway, so that light would not have been significantly delayed by bouncing off the dust on its way to Earth.

        1. Geomoetry disagrees with you. The angle of the triangle may be very narrow, but the distance from Source (point S) to reflector (point R) to Observer (point O) (distance SRO) is going to be longer than distance SO.
          Drawing ASCII art i nproportional-spaced fonts is frustrating at best, but here goes ; consider underscores “_” as space-fillers.
          You can work out the distance from S to R by the time delay for the light to reach us after we saw the outburst.
          (SRO – SO) = Delay * c
          The clouds can be in on this side of the star as well as behind it.

          1. OK, let’s look at the geometry more closely. In fact the shell of dust we see illuminated at any given moment is that for which the total light travel time from the star to the dust to Earth is the same for all points on the shell. That shape is an ellipsoid with the star at one focus and the Earth at the other.

            It’s true that a large portion of that shell is on this side of the star, but we’re seeing that portion of it nearly edge on. And no part of that shell lies directly between us and the star.

            So I think Michael’s claim is defensible that most of the dust we see lit up is behind the star. The lit-up dust on this side of the star appears to us as a narrow ring around the perimeter of the lit-up region, and the lit-up dust in close angular proximity to the star is certainly behind it.

          2. Hmmm, yes, an ellipsoid.. However it is certainly not the case that all the illuminated clouds lay behind the star’s position. Which is the way that I read the original post.

          3. True…the post doesn’t specify that the dust is mainly behind the star.

            Would we still get the observed effect if the dust surrounded the star in a uniform fashion?

          4. Pretty much, yes.
            These shells are generally either spherical, or have an axis of rotational symmetry. (See my link down-thread to pictures of Eta Carinae).
            While these are news to Jerry, they’re hardly news in the astronomical community. I remember that the ~ 1993 discovery of light echoes from SN 1987A didn’t really stir news other than because it was [big font] a light echo from the brightest SN in living memory [/ big font].
            I have a vague memory of reading about nova light echoes being caught in the early 1970s.

          5. Tracking the peak illumination of the dust results in a spherical expanding surface [only a few light weeks or light months thick] centred on the star. [I’m ignoring the apparent stretching of the sphere along our line of sight from our frame/POV]

            Now consider any one of the dust particles in the front half of this spherical surface… The photons from the burst that hit that particle will hit it’s rear mostly & very few of the photons will be deflected towards us compared with a similar particle in the other half of the sphere where we see a greater portion of the illumination of a particle.

            The particles in the sphere can each be though of as miniature moons with the “half moon” to “full moon” particles being in the hemisphere ‘behind’ the star from our perspective.

          6. Ah, the phase of the particles. Good point. I’d have to get pencil to paper to work out what relative effect that would have on the nebula’s surface brightness.

  5. I think the cloud would move. Light carries momentum. 600,000 times brighter than the sun sounds like a lot. I’m sure that all the dust particles are bombarded by a lot of photons. Some are scattered in our direction but the bulk of them aren’t. All those collisions should get that dust cloud moving.

    1. The dust is a long way away from the star. Radiation pressure is proportional to Luminosity divided by the square of the Distance, so it may be rather small.

      For example the radiation pressure of the Sun at Earth’s distance is about 9 micro-Pascals. So at Earth distance the pressure from this star would be still be only 5 milliPascals. At interstellar distances it would be much-much less.

    2. I think the cloud would move.

      It does. Because the

      Light carries momentum.

      The dust clouds have been ejected from the surface of the star (mostly ; at distance they’re being mixed into the interstellar medium, from turbulence down to viscosity) and then pushed away by the light pressure from the star.
      Re-watch the video and you’ll notice that the central star reddens significantly towards the end, while the field stars don’t change colour (therefore, it’s not a detector effect, but an effect at the star). That is probably the next shell of debris that has been ejected form the surface of the star. In time – thousands of years – this dust shell will become another diffuse light-reflecting shell.
      Look up Eta Carinae : in 1853 (IIRC ; about then) it went through a paroxysmal event, the dust cloud of which became large enough to be visible to the Hubble in recent decades.
      Take a look at http://www.star.ucl.ac.uk/~apod/apod/ap121230.html
      If that doesn’t say “ticking time bomb” to you then your astronomical straw-clencher needs re-calibrating.
      (Incidentally, the evident rotational symmetry axis MAY be the rotation axis of the central star ; or it may be the pole of the plane of the orbit of an unresolvable central pair of stars ; we don’t know if there is one star or several in the centre of this mess.)

  6. Slow light experiments (http://en.wikipedia.org/wiki/Slow_light) cause light (group velocity) to slow down in a dense atomic cloud to tens of meters / second. This is done by manipulating the allowed transitions of atoms using a pump laser beam while another beam (probe) tries to move through the atoms. Harvard, Texas A&M, and Berkeley have done this in different ways (all remarkable). Although, it is more or less invented at Stanford and a little at U. of Arizona.

    This nonlinear effect, of course, is orders of magnitude greater than the effect static index of refraction has on light propagation.

    1. Actually, in relativistic physics, it’s a bit more complicated than simply saying that these events happened in the distant past. In a very real sense, for us, they’re happening right now.

      The universe is weird, and trying to figure it out will make your head hurt. But, as they say, no pain, no gain….


      1. Good point. Saying “what it’s like right now on the other side of the observable universe” is, so I understand, something of a meaningless statement. Thinking about light cones and causality can definitely give me a headache.

        Still, thinking of what we’re seeing as the distant past has a lot of utility, since it allows astronomers to study the history of the universe.

        1. It’s not meaningless at all. It’s just a question that can’t be answered for billions of years, unless we come up with a non-local theory of sub-quantum reality that can be exploited.

          The notion that a question is meaningless if the answer cannot be determined in principle until some future time strikes me as rather silly.

          So I reject entirely the dreadful notion that distant events happen when the photons arrive here, as opposed to when they were created.

          1. You may reject it, but it is the chosen reference frame of astronomers. Because a) it is the simplest for a local observer (classically) and b) it is the only relevant for a local observer (relativistically).

            That doesn’t mean we can’t tell the history akin to how it is done in paleontology.

  7. What I am personally having a hard time getting my head around is that what we are seeing is supposed to be successively illuminated dust, as if the dust were relatively stationary as the light front sweeps over it. But as we watch one can clearly see numerous specific swirly features in the dust that really do look like they are moving outward. Could it be that the kinetic energy of the light is moving the dust? The speeds though must be incredible.

    1. Photon scattering does accelerate dust, but not enough to be visible from this far away, even in time-lapse.

      Those swirly features you’re seeing are connected to each other in extended 3D structures. What we’re seeing is a moving 2D cross-section of those structures. The illusion of motion comes from the fact that adjacent layers share common structure that changes smoothly over some spatial dimension. That smooth spatial change is transferred into the time dimension by the moving wavefront of light.

      Imagine taking an MRI scan of a cinnamon roll, and making a video of the successive slices. You’d see moving swirly patterns much like you see here, even though the roll itself is stationary.

      1. I suspect some diligent astronomer is (or already has) using this data to determine the three-dimensional structure of that dust cloud. If not the whole thing, at least the half that’s beyond the star that lit up.

        1. I’ve certainly seen work like this being discussed years ago for a continuing similar series of light echoes from SN1987A.

      2. I later started to think along the same lines. That there were extended eddies that were much longer than were being illuminated at one time. So the light front was making a series of ‘optical sections’ to borrow a term from microscopy. The information above says that the eddies were possibly formed from magnetic fields. The illusion of movement is startling, though.

  8. Things like these make comparison of Science with religions (or mythologies, lovely philosophical and ethical-teaching stories) very different. The Wonder of Reality is much much larger than the a man’s, a group of men’s,a tribe’s, a nation’s imagination.

    Vive la ciencia!

  9. I was wondering how big the illuminated part of the cloud would look from the earth. It’s 8 light-years across by the end of the video, and the star is about 20,000 light years from us. That means the cloud would subtend an angle of about 8/20000 radians from earth, or ~1.4 minutes of arc. The full moon is around 30 minutes across, so the cloud would look less than 1/20th of the diameter of the moon. Of course, that was back in 2006. It’s grown by another 14 light years by now…

Leave a Comment

Your email address will not be published. Required fields are marked *