Artists Needed for... art!

I'm not exactly an artist but I did try to render the binarysystem with a dust disk around the companion in POV-Ray (seeimage below). If you think you can use this in any way you'refree to do so. I can also provide the source file if you'reinterested.Technically, this could be rendered at any resolution although myold P4 already took 15 minutes just for the 640x480 draft. Cometo think of it you probably have a access to a powerful linux clusterfor data mining and could render it in 15 seconds ;) I can also adapt it according to your comments (as far astechnically possible) but it depends on your time frame. Ifyou reply here or email me on Monday I could do some workon this on Tuesday, but after that I won't have much timeuntil July 27th.

Here is my illustration of the epsilon Aurigae system (this has been reduced in size considerably - the original is 2000x1125):Please let me know if you would like to use it; I will provide the file in its original size.

Brian, that is truly beautiful! We would love to use it. Can you send the original size copy to Rebecca? We will be in contact with you as we prepare the press release. Also, do you have a preferred type of Creative Commons (or similar) license you'd like to apply to it? Heck, maybe we can get some posters made of this to hand out to contest winners. :)

Hi, froesccn. Since you are a 3D modeler, maybe we can use you to make some 3D models so we can create a stereo image. Do you think you can make a 3D model similar to Brian's picture below? It doesn't need to have his textures or look as pretty as his. Just try to get the structure and elements similar. If you can do that, then we could render two frames and put together a side-by-side pair and an anaglyph pair for stereoscopic viewing to show people the structure of the system. What do you think?

What a gorgeous illustration! Thanks for posting this. We will definitely want to use it - I look forward to hearing from you. Thank you!!!

While an interesting drawing the proportions are wrong. The eclipsing body is 10 times the size of the F star. From the perspective of the eclipsing body the F star should be considerably smaller (much more than 10 times smaller). As is the eclipsing body is WAY too small or the F star WAY too big. The eclipsing body is some 38 AU away from the F star. At that distance the F star would appear very small in comparison. Jeff

I originally tried a composition which used a much smaller F star. However, that made it difficult to understand what was happening in the image. The smaller body becomes lost in the image, and suddenly the eclipsing body doesn't appear to be eclipsing anything. I believe Casey Reed probably came to a similar conclusion when he produced the image that is being used on the website right now - embellishing the size the F star is a way of assisting the readability of the image.I understand the desire to be scientifically accurate, however that often comes at the expense of producing an image which is understandable (especially without prior knowledge of what the image is about). It is often essential to err on the side of readability rather than complete scientific accuracy when the goal is to clearly communicate a concept (a glance through the artwork in a copy of Sky & Telescope or Astronomy magazine will usually demonstrate this).That said, it's not too late to make adjustments. If the Citizen Sky staff would like me to reduce the size of the F star, I will certainly oblige.

The question here is what is the educational goal of the illustration? Is it more important to show scale or more important to show the structure of the system? It seems we can't do both from this viewpoint. Structure will have a bigger impact on a viewer's ability to apply the image to the data. But for participants who may want to model the system, then scale would be equally as important. Can you create a second image that shrinks the size of the primary star without creating too much extra work for you? If so, we'll use both images. Then the user of the image can choose which one they want to use in context with how they plan to use it. For the press release, we'll probably use this image for the reasons Brian mentioned. So if it is a lot of work to make a second illustration, then let's stick with this one. One reminder: The theory that the central star is an F-type is not universally accepted. It's the dominant theory right now, but it may not survive the events coming up in the next couple of years. :)

I personally prefer a scientifically accurate drawing. Otherwise to me it just become science fiction and anything goes. With astronomical scale objects getting correct proportions can be very hard. It does not have to be perfect, but should be close. I think if the primary star in the drawing was reduce to about 1/4 or even 1/8 the current size it would certainly help. When I look at the current drawing I get the impression that since we are close to the eclipsing body it is actually smaller than the primary star. That misses one of the big points of the system, that the eclipsing body is gigantic. Perhaps the largest one thing in the universe. While the primary star is a supergiant and one of the largest stars known, it is truly dwarfed by the eclipsing body. If you can shrink the primary star I think this could be a very valuable drawing.

Here is the revision - Taking Jeff's suggestions, I tried some different diameters ranging from 1/4th to 1/8th of the original size. I settled on a final size that is approximately 1/6 of the original. I hope this is a reasonable compromise between artistic license and scientific accuracy. I will send both images in their original sizes to Rebecca. Regarding a CC license, I am usingCreative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.

Hi again, just returned from vacation. Good to see you found a better illustration than mine in the meantime;-) Regarding the stereo view, I looked a bit into that and although possible it is not as simply straightforward as nudging the camera a bit, you have to consider issues such as scale(eye distance vs. astronomical distance), how the image will be viewed, where the plane of interest is, how the images overlap and so on. It also doesn't help that I usually have problems viewing stereoscopic image pairs myself anyway. Not to mention I would have to make a model with more structure. For now, I'm pursuing another benefit of a 3D model, namely the potential to animate it. This has it's own learning curve but seems promising.It actually raises some "scientific" questions: Assuming the companion is a dust disk, do we have any idea how fast the disk itself would rotate? I assume it would have some sort of differential rotation ... also, would the disk be noticably distorted by tidal forces of the primary? Furthermore, should the disk contain a hole shaped out by a binary companion, could we estimate how fast these would orbit each other?

You ask some good questions. A big part of the mystery of epsilonAurigae is we do not know much about the eclipsing body. Since it does not radiate (at least not significantly), we cannot make direct measurements.The same with the proposed binary stars within the central portion(hole). The object may consist of gas, dust, hunks of material or some combination. What created it? There certainly is a gravitational influence between the F star and unknown body, but remember the distance between them is great, something like 25 AU and they are both around the same mass (15 solar masses). Some professionals have called the eclipsing body a round paving brick with a hole in the middle. While certainly not a paving brick it does show the thinking about this mysterious object. Noone knows. The only way we know it even exists is because it passes between the Earth and the very bright supergiant F star. The timings of the eclipse provide a weath of information, but because we cannot detect the body itself, there is still a great deal unknown. Extensive spectroscopy, polarimetry and interferometry during the eclipse may provide more information to help solve this mystery. We are hoping during this eclipse we can find out more about it.

Sounds like there is still a lot of artistic freedom ;) Here is a first version of the animation, which looks something like this: You'll need to have a DivX codec to view it, though.

Woah! That was awesome! Too bad we didnt have that yesterday or I would have put it in the Lite Brite video.Maybe v2. I especially like how you got the central hole. So what is your next step? aaron

Hi Froescnn, Actually we do have some pretty good information on the rotation of the secondary from the absorption spectrum. The absorption lines in the spectrum move from being redshifted to blue shifted during the eclipse as we see first the leading and then trailing edge pass infront of the star. See the explanation of my spectroscopic observation of the start of the eclipse in the dataanalysis forum and the Lambert and Sawyer paper referenced there. http://www.citizensky.org/forum/first-publication-cbet-1885 The curve of the doppler shift as it changes throughout the eclipse is interesting as it is about +20km/sec on the leading edge and-40km/sec on the trailing edge.(Lambert and Sawyer paper, fig 2)Perhaps Dr Bob knowsmore about how this relates to the disc geometry andthe velocity profile of the material in the disc. Really impressed by the graphcs here BTW Robin

Hi Robin, thanks for the reference to the paper. Curious. Unfortunately the explanations for the velocity difference still seem to be rather speculative. But, assuming there the order of magnitude of the rotational speed is about 20km/s near the edge of a dust disk with 9 AUradius, that probably means a rotation takes close to 15 years. I'll conjecture it's exactly 13.5y which is optimal for making a cyclic animation with the 27y orbit, qed. Congrats BTW for detecting the first signs of the eclipse! Christian

Hi Aaron, glad you liked it. Next step, good question, I don't have a master plan for this. I tried to read up to refine the model and think the primary was too large initially and Inow set the radius to 1.4AU. I also increased the rotation of the disk by a factor of 4 to a period of 13.5y. Ialso went out of my way and derived a virtual light curve from the rendered frames which more or less didn't fit the bill. Of course it can't be expected to be calibrated to actual magnitudes in any way, but the dust disk contributed way to much light so the system actually got a lot brighter when the disk moved to the back maximizing its visible lit area. It could be fixed by cranking up the absorption but then the disk is essentially black on black. Also, absorbing without reemitting is probably not phsysically sound. And even then Igot a small but distinct secondary eclipse when the B5 stars went behind the primary, which Iunderstand is not observed ... there was mention that the b5 star was visibly occluded by the disk but I don't see how it would work. If I reduce the tilt so the star is hidden I also lose eclipse mid-brightening. Well Isuppose Iwon't solve the mysteries by myself and should call it a day ;) This leaves the more mundane task of wrapping it up. That depends mainly on whether you are interested in using it. I originally had intendend to show multiple viewpoints, but frankly the top view looks totally boring. I'm also not sure with a DivX AVIis the best format for publishing, maybe it would be better to find software which can output MPEG. It might be nice to embed the light curve so you can relate it with the orbital phase. Do you have a smoothed processed profile for an entire 27y period? I downloaded the data from AAVSObut of course it is still with scatter. Christian

Ok so I finally got around to wrapping this up. You can view the final animation at http://www.chrfr.de/astronomy/epsaur.html If you wish to download it it's available as Flash and MPEG: http://www.chrfr.de/animations/astronomy/epsaur.swf http://www.chrfr.de/animations/astronomy/epsaur.mpg Feel free to use it for Citizen Sky. The Flash version would obviously be suitable for embedding it on the web site (see html source of the first link above). Also, I'll remove the intermediate files from my server so some links and embedded data in the above thread will be broken.

Brian's artwork just made the front page of wired.com! The article is here: http://www.wired.com/wiredscience/2009/08/citizensky/

Brian, can I use your image on my site to organise observations for the French?

If the word "artist" is a person whose creative work shows sensitivity and imaginary wider.