Asteroid brightness variation? (APOD 05 Apr 2007)

[dlw]

Why does the apparent brightness of the asteroid 2006 VV2 change so dramatically along its tragectory? It's hard to imagine that the exposure varies that much. Does it pass through the Earth's shadow?

[BMAONE23]

http://antwrp.gsfc.nasa.gov/apod/ap070405.html
If I had to guess, I would think this would be due to surface brightness differences that are displayed while the asteroid tumbles in space.

[Chris Peterson]

Why does the apparent brightness of the asteroid 2006 VV2 change so dramatically along its tragectory?

You are seeing variation caused by changes in the surface brightness over 77 minutes of the object's 146 minute rotation rate. It is primarily through photometry on asteroids that we learn their rotation rates (although this one got close enough for radar measurements).

[Zathris]

Chris Peterson is correct. The light curve dose show the rotation of 2006 VV2's 2.43 hour rotation period. That also tells us something of the nature of the object. Unlike meny asteroids 2006 VV2 is a solid mass not a rubble pile. Since the gravitational field of a mile wide object is less then the centrifugal force of such a short rotation period.

[JohnD]

Nonsense!

Or more politely, that may not be the explanation. A continuous exposure that showed such discontinuity could have that explanation, but this is not a continuous exposure. It is described as a 'composite', many individual exposures, each elongated into a dash by the duration of that exposure, superimposed on the same star field.

John

[Chris Peterson]

Nonsense!

Or more politely, that may not be the explanation. A continuous exposure that showed such discontinuity could have that explanation, but this is not a continuous exposure. It is described as a 'composite', many individual exposures, each elongated into a dash by the duration of that exposure, superimposed on the same star field.
John

Just what are you objecting to? Light curves are almost always constructed from multiple individual exposures, since you can't take accurate measurements from a continuous trail. And in this case, I don't think the description accurately describes the combination process. I doubt the starfield is a separate shot, but rather the stars are in each image, and those images are then stacked using the stars as the position reference. the position of the asteroid in each frame is intrinsically correct, it wasn't synthesized.

It is certainly true that the variation in brightness is caused by the rotation of the asteroid. It is visible in other images, as well.

[JohnD]

Chris,
I have no doubt that a photograph could be taken by exposing the plate continuously for the length of time it took for the asteroid to travel that far. Just as a fixed camera would show 'star trails' as the objects moved, a camera attached to a telescope with a motorised mount would show fixed stars and a line of light as the asteroid moved. That would not be a "composite" as the APOD legend described this picture, but a single photograph.

We have often had the pleasure of such composites on APOD, for instance the recent occultation of Saturn by the Moon, so if this was a single photograph, why did they describe as a composite?

I also note that you are either a pro-astronomer or a very knowledgable amateur, because of the knowledge you have to be able to say that this asteroid has such-and-such a 'day', and that the picture was taken over a shorter period than the day. So I wonder if, in my ignorance, we are talking at cross purposes.

I looked again at the picture and see that the streak gets less bright towards the right hand end, as if it was dimming. THAT could represent the rotation through half its 'day' of a object with dark and light faces.
What I am talking about is the 'dashed' nature of the whole line, as in the Saturn/Moon picture, that says loud and clear "composite of many exposures"
However, I am puzzled as to why it would be necessary to take that many pics if the duration of the event was just over one hour. Or, if it was a continuous exposure, why the passage of the asteroid is shown as a dashed line at all?
Or did the APOD legend get it wrong?
Hope you can resolve my difficulty.

John

[Chris Peterson]

I have no doubt that a photograph could be taken by exposing the plate continuously for the length of time it took for the asteroid to travel that far.

Yes, such an image can be taken very easily. See http://www.cloudbait.com/gallery/comet/asteroids.html
for some examples.

But scientifically, such images aren't very useful for measuring light curves. What you really want is a lot of individual images, collected at known times, and producing ideally a small spot, or at least a very short trail. With a sequence of such images you can accurately determine the position and brightness of the asteroid at each time. With a continuous trail you can do neither.

Just as a fixed camera would show 'star trails' as the objects moved, a camera attached to a telescope with a motorised mount would show fixed stars and a line of light as the asteroid moved.

That's correct. The APOD image was also produced this way, but instead of one image, many were collected. Because the mount was tracking, the stars are seen as fixed and the asteroid is trailed in each 60 second exposure exposure.

Since any one of these images, while scientifically useful, would not be all that impressive visually, Bob summed all the images together when he was done. The stars were fixed, so they remain as points. but since the asteroid was moving, you can see a much longer section of its trail, broken just by very short gaps while the camera was downloading its data and therefore not recording. You can see the same effect in my image of this asteroid at the above referenced page. I made shorter exposures and had longer download gaps, but otherwise the same thing is going on. My total image is shorter, too, so the variation in brightness with rotation is hard to see, although I easily measured it in the individual frames.

Mathematically (and visually), the result is identical to what you would see with a single 77 minute exposure where every minute someone stuck a hat in front of the telescope for 5 seconds.

That would not be a "composite" as the APOD legend described this picture, but a single photograph.

Just so we are clear that "composite" in this case doesn't imply any sort of fudging (which I believe the lunar/Saturn image may have required). The image is nothing more than a summed stack, which is the same technique used for the majority of long exposure astronomical images. If you look at the images on my site, or any other amateur's, you'll see that they are shown with exposure times often measured in hours. But in actuality, these images are summed from subexposures that are usually between a few minutes and perhaps 30 minutes long.

I looked again at the picture and see that the streak gets less bright towards the right hand end, as if it was dimming. THAT could represent the rotation through half its 'day' of a object with dark and light faces.

That is exactly what it represents. You can't be certain of this looking at the composite alone, since you have no way of knowing whether thin clouds drifted through the field during the exposures. But we know it is true by looking at the individual subframes. That's because standard photometric methods are used comparing the asteroid's brightness to the brightness of surrounding reference stars. Anything that causes systematic dimming (like clouds) is compensated for. This is a key reason why you want lots of short exposures rather than one long one.

However, I am puzzled as to why it would be necessary to take that many pics if the duration of the event was just over one hour. Or, if it was a continuous exposure, why the passage of the asteroid is shown as a dashed line at all?

Answered above, I hope. But to summarize, you take many pictures so that you can analyze each photometrically and astrometrically to get accurate intensities and positions, neither of which is possible with a single long exposure. The gaps are simply artifacts caused by the fact that there were short periods when data wasn't being taken.

[JohnD]

Thank you Chris for such a comprehensive answer.

No 'fudging' implied! Just to relieve my ignorance.

John

[Zathris]

Nonsense!

Or more politely, that may not be the explanation. A continuous exposure that showed such discontinuity could have that explanation, but this is not a continuous exposure. It is described as a 'composite', many individual exposures, each elongated into a dash by the duration of that exposure, superimposed on the same star field.

John

Greetings all,
I will explain the process I used to create this image and it’s accompanying animation to help alleviate anyone’s misconceptions.

My intent was not to do any precise scientific measurements, but to create an animation of one of the few astronomical events that could be captured in real time that shows something other than a static event. I captured the Solar eclipse of 04/08/05 (15% at my location). I captured 4 hours and 41 minutes of the Mercury Transit on 11/08/06, and I thought I would capture a fast moving NEO glide past M81 & M82.

I was out and had my equipment setup and balanced twenty minutes before dusk. I knew that 2006 VV2 was already passing M82 as the sky was darkening enough for me to polar align and synchronize my G-11. While it was still too light to start the image run, I was framing and focusing the camera.

The asteroid was moving at a tad over 0.5 arc/sec per pixel. This would allow about 12 seconds per pixel exposure time, as I binned the ST8-XE camera at 2x2 giving me 6.18 arc/sec per pixel image scale through the Orion ED80 I shot the images through. Since my intention was the animation I chose a 60 second exposure to give a five pixel streak per exposure. This time limit was also selected because the Moon was at about 82% full and I was shooting with the clear filter.

I started the actual saved images at 03/28/07 8:15pm or 03/29/07 03:15 UT. For the next 77 minutes the ST8-XE took 60 second exposures with a 5 second gap between exposures to allow the image to download to the laptop. The ST8-XE using it’s internal guide chip guide corrected the tracking of the G-11 to within one pixel for the entire 77 exposures using one second guide exposures.

The next day I used CCDSoft and “aligned” the images. This was mostly to correct for any field rotation that may have crept in from a hasty polar alignment. Then I used CCDSoft to median combine the 71 images. This allowed me to make a background image of the star field and galaxies, but the asteroid was not present because of the median combine blending method.

I then used PhotoShop, with the FITS Liberator plugin to layer all 71 individual frames on top of the background star field image. When I imported the 71 frames into PhotoShop, I set the histogram range to 5,000 for each frame. This gave the individual frames a uniform brightness level that removed the sky illumination caused by the Moon, yet left the stars and asteroid easily visible.

The “composite” image was the background frame and the 71 individual movement frames. To make the static “flat” image I just set all of the blending methods to “Lighten”. This resulted in the background frame remaining consistent, and the asteroid for each frame showing.

The animation was just the background image and one individual frame after the other saved as a separate GIF image. These where later used to create the animated GIF file.

There where no clouds visible during the exposures. All frames where the same exposure time. All image processing was done in an exactly consistent manner. When I started assembling the image I saw the asteroid’s brightness drop in the first three or four images, and wondered if I was doing something wrong. I checked to be sure that everything was done the same way each time. When I saw it was I just continued to process till I was through. I saw the light curve that this produced but was surprised. At the time I didn’t know 2006 VV2 had a rotation period of 2.43 hours. I knew most asteroids had much longer periods.

A few days later 04/02/07 I imaged 2006 VV2 again for 3 hours and 31 minutes (the entire FOV of the ED80 and ST8-XE) when the asteroid saw moving just below 0.4 arc/sec per pixel. I wanted to see if I could see the entire light curve, and if indeed it was a light curve I was seeing. You can see the image here:

http://www.budgetastro.com/web/2006VV2- ... 9-1240.jpg

or the animation here: (beware it is a 16 meg file)

http://www.budgetastro.com/web/2006VV2-04-02-07.gif

The same procedure was used for these images as the flyby image. The only differences are that they are 60 second exposures with a 60 second gap between exposures. Also the moon was full and fairly close, so the histogram range was different (I don’t remember what I used), but consistent for all frames. This image also clearly shows the light curve of 2006 VV2. I saved the PSD files, so if anyone would like them just let me know which one or both you would like and I will upload them.

If anyone would like to know anything else I haven’t covered, just ask.

Robert Long….
Vado, NM.