Starship Asterisk*

DonJ wrote:Geck: Given that the two images I pointed out are on oposite sides of their respective focal points, close to the distorting galaxy cluster, and relatively un-smeared, does that mean the Galaxy is close to the focal point? Also, given the two separate views, can one create a stereogram and get some incliation of relative distance of stars within the Galaxy? Or are they simply too far away to get significantly different perspectives in the two views?

I'm not sure what you mean by focal point or whether it has any meaning, here. The distorting galaxy cluster is not a smooth lens; its mass greater here and there and so it's more like lumpy piece of glass, hence the wild distortions and double, triple, quadruple, etc images of background galaxies. I wouldn't expect any kind of stereogram to be possible without us having a simultaneous view a few hundred million light years offset from our position, and the lens would probably make it impossible because the distortion wouldn't be the same from that other vantage point. Kind of fun thought experiment.

neufer wrote:

---

This is actually quite a good image to illustrate gravitational lensing. The question is, if you didn't already know what a human being looks like, could you figure it out from just the reflections, not knowing the exact shape of the mirrors? The answer is formally "no", but practically "to some extent". This would involve an iterative process of identifying what appear to be common structures, using those to make tentative models of mirror surfaces, then testing different proposed "human models" with those mirror models to see if the resulting simulated images look like the observed ones. This can be done repeatedly, and we may well end up with actual models of both the mirrors and the human that resemble the real things. But this isn't something that leads to a closed solution, so we might also end up with something very different- a different mirror and a different human that produce the same image.

This is how gravitational lenses are analyzed. It's usually best with near point sources in the background, not extended galaxies. And the main interest is usually in understanding what the lens looks like, since that informs us about dark matter distribution. Reconstructing a background galaxy is somewhat interesting, but usually not of much value since we already know what galaxies look like, and knowing about one more doesn't extend our knowledge much.

Maybe if we could build an extremely large convex lens we would be able undo the gravitational lensing? Of course we would need a giant optician to build one for every different lensing event. That wouldn't work.

Fred the Cat wrote:Maybe if we could build an extremely large convex lens we would be able undo the gravitational lensing? Of course we would need a giant optician to build one for every different lensing event. That wouldn't work. :no:

A gravitational lens is neither convex nor concave. The thing that makes it impossible to "undo" a gravitational lens with complete certainty is that it distributes light in an ambiguous manner. There are multiple possible objects that can produce the same image.