APOD: V838 Light Echo: The Movie (2014 Jun 17)

[JohnD]

Was the 'movie' compensated for the delay of reflections from gas/dust clouds behind the star?

The 'shell' of light radiated by the flash was in all directions. So what we see is a composite of light reflected from the surrounding clouds, up to four light years from the star, as it was seen from 2002 to 2006. Light reflected from clouds at the same distance from us as the star would have arrived up to four years after the direct flash light, and from clouds behind it, up to eight years later.

John

[Chris Peterson]

It's likely that all the material we see in this nebula was ejected from the progenitor star.

But then, those material should travel faster than light, because it seems to have been already there.

That assumes the flareup that produced the light echo was this star's first rodeo. Unstable stars tend to go through multiple outbursts, so some material was probably released previous to the bright flash that produced the echo.

[geckzilla]

I'd love to see a simulation of light traveling through an irregularly shaped medium like this to help my brain cope with what the pictures show. Even the still frames are somewhat baffling to me.

[MarkBour]

I'd love to see a simulation of light traveling through an irregularly shaped medium like this to help my brain cope with what the pictures show. Even the still frames are somewhat baffling to me.

Yes, that'd be nice. The shape should be something like neufer's diagram showed. We know that all of the points that are brightly lit up by this in a Hubble image taken in year 2002+n took n more years to travel to us than the initial 2002 flash. So you can plot those points on a path that must be n ly longer than a straight line. This is kind of a ray-tracing job then to work back to the actual shape of the cloud particles that we have seen.

But there seem to be more unknowns than I like here, which would prevent definitive solution. If we rely on some independent measure of the distance to the nebula, then we could use that to solve for what angle it is from the emission point to a given piece of cloud that the light bounced off of n years behind the initial flash. Even then, I'm not sure if that does us much good at reconstructing much of a map of the cloud, unless we know if the light is merely reflected, or is absorbed and re-emitted. In the latter case, I think it might give us a picture of all of that dust, but in the former, lots of dust may block light rather than reflect it (or reflect it away from us, which is about the same thing to us) and all of that uncooperative dust would appear as darkness, and our reconstruction of the cloud would be hopelessly incomplete.

I agree with nitpicker who pointed out this has to work from front to back, so, as he realized, it should eventually shrink back down to a point. If he's right, that will be great to watch.

[tetrodehead]

Please provide a link to the original 4 frames. Thanks.

[geckzilla]

Please provide a link to the original 4 frames. Thanks.

There are more than four at this point but here are the first four:
http://www.spacetelescope.org/images/heic0304a/

[geckzilla]

Mark, I don't think we need an exact simulation of this same structure. A roughly equivalent one demonstrating the concept would help a lot even if it doesn't look just like V838 Mon.

[Nitpicker]

I don't find the smoothing or morphing between frames makes it any more difficult to visualise. Either way, the pattern gives the illusion of dust movement. The illusion is real, the dust movement is not.

But I don't really understand neufer's paraboloid sections analogy, either.

I have to imagine it as follows ... Each particle of dust is a certain distance in light years away from the star. As each particle is illuminated by the starburst, it becomes a secondary source of reflected light. The time since 2002, when we see each new particle illuminated, represents the extra distance the light has to travel to Earth, compared with the direct distance from star to Earth. So, all the dust in the column within the direct line of sight of the star, was illuminated to us on Earth at the exact same time as we first observed the starburst. And as the total light paths from star to dust particle to Earth become longer, we progressively see the rest of the dust. (And yes, this is simplifying matters somewhat by ignoring internal reflections from particle to particle.)

[Chris Peterson]

I don't find the smoothing or morphing between frames makes it any more difficult to visualise. Either way, the pattern gives the illusion of dust movement. The illusion is real, the dust movement is not.

But in the morphed version, the actual movement is made real. You can see material in one frame being interpolated outwards in the next frames. That's simply wrong. It's not an illusion in the video. In the still frames you can find structure in each frame and see that it doesn't move.

[geckzilla]

I think we are on the same page, Nit. It's just that the picture conflicts so much with how I imagine it versus how it really looks that I am having difficulty reconciling the two. Have a look again at RS Puppis's light echo. This one I have no problems with at all. RS Puppis is so much easier to comprehend. Its nebula is not partially invisible and has multiple light pulses flowing through it. This also seems to be more of a side view rather than a spherical volume.

[Nitpicker]

I don't find the smoothing or morphing between frames makes it any more difficult to visualise. Either way, the pattern gives the illusion of dust movement. The illusion is real, the dust movement is not.

But in the morphed version, the actual movement is made real. You can see material in one frame being interpolated outwards in the next frames. That's simply wrong. It's not an illusion in the video. In the still frames you can find structure in each frame and see that it doesn't move.

I don't see what you are seeing. I don't know what you mean by "the actual movement is made real". There is no actual movement. We are just seeing different parts of the dust structure at different times. To me, the difference is a bit like looking at two representations of the visible spectrum, one smooth and continuous from red to violet, and the other discretised into ROYGBIV colours. I'm not saying you are wrong, just that the morphing doesn't make it any better or worse for me, as a visualisation aid. (I do spend a lot of my professional life looking at sections moving variously through 3-D models, however.)

[Chris Peterson]

I don't see what you are seeing. I don't know what you mean by "the actual movement is made real". There is no actual movement.

I disagree. You can see actual dust and gas structures moving radially outwards in the morphed video. You do not see this in the individual frames. The morphing process has added a major artifact (not surprising- I've never seen morphing software that wouldn't assume movement, and build it into the interpolated frames).

[Nitpicker]

I don't see what you are seeing. I don't know what you mean by "the actual movement is made real". There is no actual movement.

I disagree. You can see actual dust and gas structures moving radially outwards in the morphed video. You do not see this in the individual frames. The morphing process has added a major artifact (not surprising- I've never seen morphing software that wouldn't assume movement, and build it into the interpolated frames).

I interpret that merely as how the interpolation between "real" frames, estimates the appearance of the echo as it moves through the structure in between the "real" frames. Interpolations always require the creation of data. What you call an artefact in this case, I would be more inclined to call a feature of the interpolation. But both descriptions are true.

(And I just "got" the paraboloid sections analogy, neufer, thanks. The vertex of the ever widening paraboloid moves progressively further behind the star with respect to our line of sight [at a speed of 0.5*c, if my brain is working correctly]. I have not yet convinced myself that the shape is parabolic, but if it isn't, it should be close.)

[Nitpicker]

I have not yet convinced myself that the shape is parabolic, but if it isn't, it should be close.

Doh, the echo shape, as observed from Earth at any particular moment, is ellipsoidal (more precisely, prolate-spheroidal), just like it says in the explanation of the APOD.

The outbursting star and the Earth form the two fixed foci of the (highly eccentric) ellipsoidal echoes, which expand over time. Based on a distance between foci, 2f, of 20,000 light years, the parameters of the ellipsoidal echo we observe over time, would be as follows:

[Nitpicker]

And here is a side view of the star (small orange cross), showing the cross-sectional shape of the light-echo contours in red, as they appeared from Earth (off to the left), for the first six years after the outburst (the most eccentric ellipse shows the shape after one month). The brown circle has a radius of three light years.

(But I'd still like to see a 3-D model of the dust cloud.)

[DavidLeodis]

Does anybody know of a link to a set of individual images of V838 Mon covering most or all of the years from 2002 to 2006 inclusive that is covered by the movie? I thought there had been such a set with APODs that have used V838 Mon as their subject, but if so I have been unable to find a set.

[MarkBour]

@nitpicker. I think your diagram is great. If I understand it, the parabolas in your diagram show a 2-D indication of 3-D paraboloids (perhaps that's what they are) of points where: IF there was a reflective bit of dust there, it would be lit up in an image at the same time. And you could plot from that and the frames from Hubble where the dust actually was that was being seen in each image. Unfortunately, I guess that's the best we can do, we only have small bits of the information of the contour of the cloud that the flash is illuminating for us. Perhaps if we had an image every day for a few years, we could map out the cloud's surface in detail. One thing I'm wondering is whether or not that actually would define the surface, or whether there are still too many unknowns left. For example, a dark spot simply means that there is more dust in front of the cloud, not that there is no dust there. Indeed, perhaps I should re-phrase my statement above to the converse: "IF we see a light spot there, THEN there is some dust there on your paraboloid for that time." The logical "IF" in the other direction is not reliable.

Of course if your diagram is used too exactly, then it would only allow us to see the thinnest rings as the echo expanded, not good-sized bands. So I'm not understanding that aspect yet.

I guess, if we had those many intervening frames we would see the echo moving quite differently than the interpolation software rendered it, and it would support Chris' point that the appearance of movement of the pieces is an artifact of the interpolation (although our own brains interpreting what they see is partly to blame).

[Nitpicker]

Paraboloids (e = 1) are not too different from ellipsoids (0 < e < 1), at least at the pointy end when e is almost 1, but these are ellipsoids. In reality, the ellipsoid echoes would have a thickness of a few light-weeks, corresponding to whatever was the actual duration of the original star burst.

And yes, internal reflections and absorptions also have an effect, but I'd think they would be second order. The ellipsoidal contours represent the spatial locations of the first order reflections from the star burst, at different times.

Also, in the very early stages of the star burst, the echoes illuminate all the dust immediately in front of the star, so if is difficult to gauge the distances accurately. This problem resolves itself later, as the echo widens and moves away through the dust.

[MarkBour]

Ah, yes. I get it, you're right, of course. Ellipsoids.

[geckzilla]

Not to be confused with footballoids or rugbyballoids.

[neufer]

Not to be confused with footballoids or rugbyballoids.

The ellipsoid in question here is much closer to a javelinoid.

[Beyond]

Hmm... looks like all the 'oids are coming home to roost.

[MarkBour]

It would be awesome, if for some light echo as dramatic as this one, we could get the "real movie". If a space telescope had the time to take a picture a week or so for 10 years. Of course I don't know if it would do humanity any good (so who would ever fund that kind of telescope time), but I still think it would be awesome. This would be a tomographic reconstruction problem.

(I hope they don't do any tomography on my 'roids )

[geckzilla]

I don't find the smoothing or morphing between frames makes it any more difficult to visualise. Either way, the pattern gives the illusion of dust movement. The illusion is real, the dust movement is not.

But in the morphed version, the actual movement is made real. You can see material in one frame being interpolated outwards in the next frames. That's simply wrong. It's not an illusion in the video. In the still frames you can find structure in each frame and see that it doesn't move.

I saw this attempt at using morphing software to interpolate data between frames of a rotating view of comet 67P. I think it's a very good example showing what kind of false movement morphing software can give to things. This is much easier to understand since we know the comet does not bend and flex like this.
https://www.flickr.com/photos/105035663 ... otostream/

[Chris Peterson]

I saw this attempt at using morphing software to interpolate data between frames of a rotating view of comet 67P. I think it's a very good example showing what kind of false movement morphing software can give to things. This is much easier to understand since we know the comet does not bend and flex like this.
https://www.flickr.com/photos/105035663 ... otostream/

Ouch.