This NYTimes video shows what I think is a series of real Hubble images which have been "morphed" artistically to appear to be continuously expanding. (also see The V838 Monocerotis Star Still Has Astronomers’ Heads Exploding) If I understand correctly it is the light-echo, not actual material that is doing the expanding at this rate. They also give four discrete images available in 2006 (shown below) and you can find a link to a video there which also seems to be made of just four images.

This is (very roughly speaking) tomography of the cloud - the parabolic "shell of reflection" expands and each image highlights a view of the material in that shell. That seems to be what the diagram illustrates.

my question is: why weren't there more images taken of this - technically speaking - awesome event, enough to make an animation without the artistic "morphing"?

note: I'm not asking "why do you think there weren't...". I'm looking for the actual reason. Are these "found images" - nobody thought of doing it and later just pulled these from archives, or where these thought to be sufficient? Are there more images available?

Here are some links to video I have found

YouTube: "Hubble: Timelapse of V838 Monocerotis (2002-2006)"


...which can also be seen at spacetelescope.org


... and this Wikipedia article links to this video


and here are some "stills"

Images from NYTimes

enter image description here

  • $\begingroup$ A newer, related question $\endgroup$ – uhoh May 5 '16 at 7:40
  • 1
    $\begingroup$ +1 interesting :) $\endgroup$ – Max0815 Feb 28 at 3:53

To get time on the Hubble Space Telescope is a difficult, peer-reviewed process. The time is usually about ten times oversubscribed. The scientific justification to get more images must be more than "it would make a nice movie".

The submission of proposals for HST time is a confidential process; so I don't know whether the principal investigators for these images asked for more at different epochs and were turned down, or whether they thought this many images were sufficient for the science they wanted to accomplish.

There is no technical reason why they couldn't have obtained more images. In fact more images were obtained (see this link to the Hubble archive) at later epochs between 2005 and 2012. A problem I guess with incorporating these images into the movie sequence were that the light echo gets fainter, more diffuse and starts to exceed the size of the detector, which would necessitate more than one pointing. They were also taken with a different camera, which might raise aesthetic compatibility problems. I show below a fairly crude colour composite I extracted from the archive for an observation in 2007.

V838 Mon in February 2007

  • $\begingroup$ Apparently your criterion: "The scientific justification to get more images must be more than 'it would make a nice movie'" was indeed satisfied. Being a transient event, time on Hubble was from the Director's Discretionary Time. See this answer. $\endgroup$ – uhoh May 5 '16 at 6:53

The V838 Monocerotis expansion (not a supernova) and the observation of the subsequent "spectacular" light echo was quite a notable event! From Nature 422, 405-408 (27 March 2003)

Nature Coverenter image description here

From Astronom. J. 135, 2, 2008 or ArXiv

".Galactic light echoes are extremely rare. The only other known example of extent similar to that of V838 Mon was the echo produced by Nova GK Persei 1901 (Kapteyn 1902; Perrine 1902; Ritchey 1902). Following early misunderstandings, light-echo geometry was properly described by Couderc (1939), and more recent discussions are given by many authors, including Chevalier (1986), Felten (1991), Sparks (1994), Sugerman (2003), and references therein".

It was the sole topic of an international conference photo from here:


From here it's noted that the Hubble observations were inserted into the observing schedule using the Director's Discretionary time, since the peer review process is too slow to accommodate observations of transient events. While there is no mention of any reason why observations were not more frequent, nor continued in 2003, one can speculate.

"Based on the highly structured appearance of the initial ground-based images, our team proposed for Director’s Discretionary (DD) time on HST for a program of direct imaging and imaging polarimetry. The team members are as follows: S. Starrfield (Arizona State University); Z. Levay, N. Panagia, W. Sparks, B. Sugerman, R. White, and myself (STScI); A. Henden (AAVSO); M. Wagner (University of Arizona); R. Corradi (Issac Newton Group); U. Munari (Padova University); L. Crause (SAAO); and M. Dopita (ANU)".

"We received HST observing time at five epochs in 2002 through DD allocations: April, May, September, October, and December. All of the observations were made with the Advanced Camera for Surveys (ACS), which had been installed in HST during SM3b in March 2002. I need not emphasize to this audience how extraordinarily unfortunate it is that no HST observations were obtained during 2003—the loss of this opportunity is truly incalculable. However, the echoes were imaged twice in 2004 through the Hubble Heritage program, in February and October. More happily, the HST Cycle 14 allocation committee did award our team observing time for an intensive HST imaging campaign from October 2005 to January 2006, and we also have two more epochs of observations scheduled in Cycle 15 for late 2006 and early 2007".

Figure 2 of the Nature paper describes the preservation of the actual light curve (history) within the structure of the light-echo shell:

enter image description here

"FIGURE 2. HST images of the light echoes The apparently superluminal expansion of the echoes as light from the outburst propagates outward into surrounding dust is shown dramatically. Images were taken in 2002 on 30 April (a), 20 May (b), 2 September (c) and 28 October (d). Each frame is 83" times 83"; north is up and east to the left. Imaging on 30 April was obtained only in the B filter, but B, V and I were used on the other three dates, allowing us to make full-colour renditions. The time evolution of the stellar outburst (Fig. 1) is reflected by structures visible in these colour images. In b, for example, note the series of rings and filamentary structures, especially in the upper right quadrant. Close examination shows that each set of rings has a sharp, blue outer edge, a dip in intensity nearer the star, and then a rebrightening to a redder plateau. Similar replicas of the outburst light curve are seen propagating outwards throughout all of the colour images."

Again from Astronom. J. 135, 2, 2008 or ArXiv

enter image description here

Figure 2. Images representing the degree of linear polarization, p, for each of the four epochs of data shown in Figure 1. Image scales and orientations are the same as in Figure 1. The image stretch is linear, ranging from black representing zero linear polarization to full white representing ~50% linear polarization. These images illustrate the apparent outward motion of a ring of highly polarized light in the light echo.

Abstract Following the outburst of the unusual variable star V838 Monocerotis in 2002, a spectacular light echo appeared. A light echo provides the possibility of direct geometric distance determination, because it should contain a ring of highly linearly polarized light at a linear radius of ct, where t is the time since the outburst. We present imaging polarimetry of the V838 Mon light echo, obtained in 2002 and 2005 with the Advanced Camera for Surveys on board the Hubble Space Telescope, which confirms the presence of the highly polarized ring. Based on detailed modeling that takes into account the outburst light curve, the paraboloidal echo geometry, and the physics of dust scattering and polarization, we find a distance of 6.1 ± 0.6 kpc. The error is dominated by the systematic uncertainty in the scattering angle of maximum linear polarization, taken to be θmax = 90° ± 5°. The polarimetric distance agrees remarkably well with a distance of 6.2 ± 1.2 kpc obtained from the entirely independent method of main-sequence fitting to a sparse star cluster associated with V838 Mon. At this distance, V838 Mon at maximum light had MV sime −9.8, making it temporarily one of the most luminous stars in the Local Group. Our validation of the polarimetric method offers promise for measurement of extragalactic distances using supernova light echoes.

  • $\begingroup$ A newer, related question $\endgroup$ – uhoh May 5 '16 at 7:40
  • $\begingroup$ But the observations were continued after this sequence. This does not provide an answer to your own question. If the scientific case to do so had been stronger,then more observations would have been obtained - unless you can find another reason. "note: I'm not asking "why do you think there weren't...". I'm looking for the actual reason". $\endgroup$ – Rob Jeffries May 5 '16 at 8:12
  • $\begingroup$ @RobJeffries It's as close as I can get at the moment. 2002: discretionary, 2003: zip - nada, 2004 and after: "the normal way." Citation instead of speculation - it's at least a step in the right direction, still working on it - not accepting it yet either of course. I'll appreciate any help you can provide! $\endgroup$ – uhoh May 5 '16 at 8:51
  • $\begingroup$ @RobJeffries If you are familiar with the Hubble, it's possible you might be able to help here also. $\endgroup$ – uhoh May 5 '16 at 8:57
  • $\begingroup$ Directors discretionary time is applied for, by astronomers. Your press cuttings simply support my argument that they applied for more images but were not allocated the time. Discretionary time is simply a small fraction that can be applied for to follow things that pop up unexpectedly. It is still judged by science merit. $\endgroup$ – Rob Jeffries May 5 '16 at 9:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.