Our skewed perception of galaxies

[Veritas2]

Our perception of what a galaxy looks like is inherently skewed given that the immense size of the galaxy itself provides a lopsided picture. If viewed from any angle other than directly head on, the closest point of the galaxy will be a much more recent picture than that of the far end, by tens to hundreds of thousands of years depending on the size of the galaxy and the angle of our view. In the example below, I illustrate this idea using arbitrary distances of a few light years difference, with the "o" representing the earth, and the "/" showing the angle of the galaxy in relation to us. The light from the center point of the galaxy takes 7.5 light years to reach earth. At the closer end however, it only takes 5 light years. And the farthest end takes 10. What this means is that the picture we see of the galaxy is actually a sort of time illusion, showing us what parts of the galaxy looked like over a period of 5 years, working back in time as we move farther away.- - - - - - - - - - - - - - 10 ltyrs.- - - - - - - - - - - - - -/
- - - - o- - - - - - - - - - - 7.5 ltyrs. - - - - - - - - - - -/
- - - - - - - - - - - - - - -5 ltyrs. - - - - - - - - - - - - /


We can further compound this skewing with the fact that galaxies are not static, but constantly moving. So with a steep enough angle we could be seeing a galaxy as it existed over 100 thousand years.

 

[russ_watters]

If a galaxy is 10 billion years old, what does plus or minus 50,000 years matter?

 

[Veritas2]

It was just the thought that this phenomenon gives an interesting view into a short term time frame of a galaxy's lifespan. We can actually see with one picture what a galaxy went through as it progressed over thousands of years.

 

[Chronos]

There is no statistically significant skew in the distribution of rotation axes of catalogued galaxies in the observable universe [source = LEDA]. Cite sources that suggest otherwise, Veritas.

 

[LURCH]

I think Veritas's statement is too simple and self-evident to require proof. If a gallaxy is 100,000 ly across, and we look at it edge-on, then the image of the edge closest to us is 100,000 yr younger than the image coming from the far edge.

A very obvious observation, Veritas, and one I never thought of before. As has been mentioned, the distance between galaxies is so great that people just tend to ignore the distance across them. I guess I always think on a cosmic scale when thinking of "galaxies" in general, and on that scale any individual galaxy occupies a space that, for all practical purposes, amounts to a single point.

 

[Mike S.]

Interesting. Has this been used to measure (on a statistical basis) how quickly the spiral arms of galaxies move relative to the underlying stars? (You'd expect them to bunch up on the side of the galaxy moving away from you and spread out on the side moving toward you)

 

[LURCH]

I don't think it could be used to examine the contraction or expansion of the stars in a spiral arm, as that change is a result of the speed at which the arms move toward or away from the observer. It could, however, reveal something about the evolution of that galaxy as a whole. If we see a galaxy with a lot of Sol-like stars on the far edge, and a lot (about the same percentage) of red giants on the near edge, then we could say with some confidence that we are looking at two different eras in that galaxie's history; a time in the near past when all the giant stars were exhausting their feul, and the "present" (or much nearer past), in which they have all gone huge. We might even be able to watch for the aftermath of some frequently-occuring event of interest (on the far edge), and predict that we could see the same event happen to stars on the near edge.

 

[Janus]

I don't think it could be used to examine the contraction or expansion of the stars in a spiral arm, as that change is a result of the speed at which the arms move toward or away from the observer. It could, however, reveal something about the evolution of that galaxy as a whole. If we see a galaxy with a lot of Sol-like stars on the far edge, and a lot (about the same percentage) of red giants on the near edge, then we could say with some confidence that we are looking at two different eras in that galaxie's history; a time in the near past when all the giant stars were exhausting their feul, and the "present" (or much nearer past), in which they have all gone huge. We might even be able to watch for the aftermath of some frequently-occuring event of interest (on the far edge), and predict that we could see the same event happen to stars on the near edge.

100,000 yrs isn't really long enough to show any significant evolution of the Galaxy. Consider that the life time of a typical star is measured in billions of years. If we take 4 billion yrs as an "average" lifetime of a star, then 100,000 yrs compares to this like 16 hrs compares to an average lifetime of a human being. If I take a sample of two groups of the same populace, with the samples taken 16 hrs apart, this isn't going to help me in uncovering as birth/death trends for the populace as a whole.

 

[Mike S.]

It takes 250 million years for the Sun to circle the Milky Way, and its orbit has a diameter of 52,000 light-years. So if you looked at the Milky Way edge-on, the Sun would appear to lag behind in its orbit by 1/5000 of its total motion. By extension, you'd see 4999/10000 of the stars on the side where the stars are coming toward you in the galaxy, and 5001/10000 of the stars on the side where they're going away. Now if the spiral arms are fixed among the stars, obviously you should see 1/5000 more spiral arms on the side going away from you. But if they move faster than that, you'll see more than 1/5000 more arms on that side. Now I'll admit this seems a little absurd to measure, especially since you can't really do it if the galaxy IS edge on, but only if it's tilted at an angle with an even smaller difference; nonetheless, there are a whole lot of galaxies out there with high-quality photos archived, and if you had a good quality image recognition program I'd think it would have a chance of pulling out a real result.