Light lines from stars - artifacts or significant?

[Delysid23]

In many astronomical images, stars have those characteristic lines of light coming out of them, usually 4 major lines forming a cross centred on the star, though not always perfectly vertical. Sometimes additional smaller fainter lines emanate from the star between the big 4 lines. I'm sure everyone knows what I mean, as cumbersome as it is to describe in words. Here's a good Hubble example:


(click to see larger version)

Now I have always assumed these lines are optical artifacts that have something to do with the physics of light and glass lenses. But suddenly I ask myself -- with modern technology and a gazillion dollar highly advanced instrument like Hubble, wouldn't there be ways to eliminate such artifacts, which though pretty in a Christmas tree decoration kind of way, are basically noise imposed on the image and wiping out a certain amount of visual data?

So either these radiating star light lines are an optical artifact of exceptionally stubborn persistence so inherent to the physics of telescope optics that they simply cannot be eliminated;

OR

They are not actually artifacts produced by lenses or other properties of telescopes at all, and instead are real visual features of stars shining through space, and if we were floating in space or sitting astride the Hubble sattelite orbiting Earth, and we fixed our beady 20/20 vision eyes free of any kind of artificial lens upon the same stars, and were genetically modified to have superhuman eyeballs with telescopic powers, those same lines would be there.

In the latter case, the lines may still be artifacts, but ones produced naturally by stars shining far away in space, rather than by our viewing apparatus. Which would mean they are still "noise". Or do they actually have some significance?

If they are optical artifacts produced by our telescopes, why can't we get rid of them in this day and age with all our newfangled technology? Surely a civilization so advanced that at any hour of the day or night I can watch a video on my computer screen showing a cat playing a piano should be able to somehow filter out those sparkly little star light-lines? Or are they beyond our reach, existing "out there" like the glowing eyes of Jesus or the dazzling rhinestones on Elvis's trousers?

[Dr. Skeptic]

The lines are caused by the supports for the secondary mirrors, decreasing the number or size of these supports will decreases the stability of the secondary mirror, equating to the ability to maintain a steady focus especially while tracking long exposures.

[iamlucky13]

If they are optical artifacts produced by our telescopes, why can't we get rid of them in this day and age with all our newfangled technology?

Get rid of them?!? No way! They look to cool.

I'm sure we can get rid of them without sacrificing accuracy a couple different ways, such as post-processing, but I'm not aware of any significant scientific harm in letting them be.

[BMAONE23]

Another way to eliminate them would be to suspend the central mirror by imbeding it in the center of a circle of perfectly clear smooth thin flat glass rather than on posts.

[Pete]

The lines you're talking about are called diffraction spikes if they're produced by mirror supports. CCD saturation produces somewhat similar-looking artifacts called saturation trails: http://www.sc.eso.org/~ohainaut/ccd/

P.S. Delysid23, I'm in Vancouver for the summer going to see the Symphony of Fire starting this Wed?

[Delysid23]

I found another excellent concise explanation of these diffraction spikes:

http://spider.ipac.caltech.edu/staff/ka ... tion2.html

I also realized that there is one bit of useful information that these diffraction spikes convey: in shots of extragalactic phenomena, diffraction spikes distinguish stars of our own galaxy from the rest of the glowing space objects contained in the image.