5 ms·

My exceptionally naive understanding is that "how long ago" and "how far away" aren't really things that cosmologists talk about. Because of relativity and the

My exceptionally naive understanding is that "how long ago" and "how far away" aren't really things that cosmologists talk about. Because of relativity and the expansion of space neither really has a meaning. It's like the idea of common time: when did something happen on Alpha Centauri? You can't really answer that. Light cones and spheres of influence and so on. "When" is only a valid question when the thing you're asking about isn't relatavistically far away. Deep stuff. Source: I just read Carlo Rovelli's The order of time. Would be overjoyed if someone who knows what they're talking about would correct/enhance this response. :-)

pdonis 4y ago
> "how long ago" and "how far away" aren't really things that cosmologists talk about. In actual textbooks and technical papers, like this one, yes, that's correct; they don't talk about them because they're not directly observable, and deriving them from the things that are directly observable is model dependent. So they prefer to talk about the things that are directly observable, of which the key ones are redshift, apparent luminosity, and (for objects that are large enough, like galaxies) angular size. The relativistic effects you mention are additional reasons why those things are not usually talked about in technical writings. Pop science writings are another matter: many cosmologists do talk about how long ago and how far away in those contexts without taking the time to add the caveats about how we don't directly observe those things.
- px43 4y ago
  I wonder how often things we observe are actually farther away than they seem, they're just rapidly moving towards us relative to their surroundings, or vice versa. Seems like it's totally possible that something could be 4 billion light years away, or 8 billion light years away, and have the same red shift. Maybe sometimes gravitational lensing would help determine if a star is behind or in front of another? I wonder if it's pretty common to observe objects that are more blue shifted than something that is clearly closer to us.
  - antihipocrat 4y ago
    I'd love to get an informed answer to this too. Is it more accurate to interpret inferred distances derived from redshift as an average estimate? If so it would be great if popular science reporting also included the confidence intervals
    - pdonis 4y ago
      > Is it more accurate to interpret inferred distances derived from redshift as an average estimate? We don't estimate distances just based on redshift. We use all three of the observables I mentioned--redshift, apparent luminosity, and angular size--as inputs to estimate parameters in our cosmological models, and one of those parameters is the scale factor--the "size" of the universe, which is what is used in the model to ultimately get actual distances--as a function of time. The models of course have error bars--no model is ever exact--and some of those error bars are due to variations of the motions of individual galaxies from the idealized cosmological expansion in the models. So in that sense distances given by the models are a sort of estimate, yes, based on the average motion.
    - mr_mitm 4y ago
      The effect you are talking about is real, but it's generally small. The cause is called peculiar velocity, and it could introduce uncertainties of a few percent, which is relatively small for cosmological standards. We know how velocities are distributed inside galaxy clusters. It leads to the "finger of god" effect, which makes it appear as if objects would be distributed such that they tend to align along our line of sight. See: https://en.wikipedia.org/wiki/Peculiar_velocity https://en.wikipedia.org/wiki/Peculiar_velocity
  - pdonis 4y ago
    > Seems like it's totally possible that something could be 4 billion light years away, or 8 billion light years away, and have the same red shift. In theory, of course this is possible. But in our actual universe, that's not what happens. In our actual universe, there is a close relationship between how far away something is and its redshift. Or, to phrase it in terms of direct observables, in theory we could have an object that was 4 billion light years away and an object that was 8 billion light years away, both with the same redshift--but they would be very different in terms of other observables like apparent luminosity and angular size. So in the general case, in theory, we would not expect any particular relationship between those variables. But in our actual universe, there is a close relationship between all three of those variables: redshift, apparent luminosity, and angular size all vary in concert and are closely correlated. That kind of observation is what leads cosmologists to conclude that our universe is expanding. > I wonder if it's pretty common to observe objects that are more blue shifted than something that is clearly closer to us. It's not only not "pretty common", it never happens except for objects very close to us, cosmologically speaking--for example, the Andromeda galaxy is somewhat blueshifted, indicating that it is moving towards us, while the Magellanic Clouds, which are closer (they are satellite galaxies of our own galaxy), are not, AFAIK, blueshifted. But all of these objects are in the same local, gravitationally bound group as our galaxy, so their relative velocities don't tell us anything useful about the large scale behavior of the universe.
elefanten 4y ago
Yes, in my 5min review it came up very quickly that “distance” and “light years” have limited uses in astrophysics. In the abstract, they reference the “z” variable of redshift in light (that has traveled for billions of years during which cosmic expansion continued and expanded the original distance) and kiloparsecs (referring to the much closer span of Milky Way through which the emissions traveled, from which they are drawing inferences about dust density therein). Wild, cool stuff!
andrewflnr 4y ago
It's perfectly possible to say when things happened at Alpha Centauri. Relativity doesn't really start messing with the notion of simultaneity until you're in a different reference frame, i.e. moving at a different speed from the observer you're comparing yourself to. But if the only difference is distance, and you know the distance, you really can just run the speed of light backward and get a meaningful time in your own past for a distant event you're just now getting the light from. Think of it this way: since you're moving the same speed, there's no time dilation between your clocks, so you can safely line up ticks between your clock and a hypothetical clock far away on Alpha Centauri or whatever. I assume things get a little more interesting at big redshifts, because you really are moving at relativistic speeds wrt what you're observing. I don't know off hand how they handle that. I will say I'm pretty sure the reason cosmologists tend to talk in redshifts is that redshifts can be pretty much directly measured and actual distance is... tricky. If you haven't already looked into the "cosmic distance ladder" you're likely to find it interesting. Source: we covered this stuff in a college class. My math is very rusty, but I did do the math.
- galaxyLogic 4y ago
  > if the only difference is distance, and you know the distance, But are we not moving in relation to Alpha Centauri at a fabulous speed, because all the galaxies are moving one way or the other and rotating as is Earth moving around the Sun and rotating around its axis?
  - andrewflnr 4y ago
    No, Alpha Centauri and Sol are both in orbit, in the same direction, around the Milky Way. So I mean, the relative speed between us and Alpha Centauri is probably pretty big compared to a car, but it's just noise in our orbital velocities, which are still peanuts compared to c.

pdonis 4y ago

> "how long ago" and "how far away" aren't really things that cosmologists talk about. In actual textbooks and technical papers, like this one, yes, that's correct; they don't talk about them because they're not directly observable, and deriving them from the things that are directly observable is model dependent. So they prefer to talk about the things that are directly observable, of which the key ones are redshift, apparent luminosity, and (for objects that are large enough, like galaxies) angular size. The relativistic effects you mention are additional reasons why those things are not usually talked about in technical writings. Pop science writings are another matter: many cosmologists do talk about how long ago and how far away in those contexts without taking the time to add the caveats about how we don't directly observe those things.

px43 4y ago
I wonder how often things we observe are actually farther away than they seem, they're just rapidly moving towards us relative to their surroundings, or vice versa. Seems like it's totally possible that something could be 4 billion light years away, or 8 billion light years away, and have the same red shift. Maybe sometimes gravitational lensing would help determine if a star is behind or in front of another? I wonder if it's pretty common to observe objects that are more blue shifted than something that is clearly closer to us.
- antihipocrat 4y ago
  I'd love to get an informed answer to this too. Is it more accurate to interpret inferred distances derived from redshift as an average estimate? If so it would be great if popular science reporting also included the confidence intervals
  - pdonis 4y ago
    > Is it more accurate to interpret inferred distances derived from redshift as an average estimate? We don't estimate distances just based on redshift. We use all three of the observables I mentioned--redshift, apparent luminosity, and angular size--as inputs to estimate parameters in our cosmological models, and one of those parameters is the scale factor--the "size" of the universe, which is what is used in the model to ultimately get actual distances--as a function of time. The models of course have error bars--no model is ever exact--and some of those error bars are due to variations of the motions of individual galaxies from the idealized cosmological expansion in the models. So in that sense distances given by the models are a sort of estimate, yes, based on the average motion.
  - mr_mitm 4y ago
    The effect you are talking about is real, but it's generally small. The cause is called peculiar velocity, and it could introduce uncertainties of a few percent, which is relatively small for cosmological standards. We know how velocities are distributed inside galaxy clusters. It leads to the "finger of god" effect, which makes it appear as if objects would be distributed such that they tend to align along our line of sight. See: https://en.wikipedia.org/wiki/Peculiar_velocity https://en.wikipedia.org/wiki/Peculiar_velocity
- pdonis 4y ago
  > Seems like it's totally possible that something could be 4 billion light years away, or 8 billion light years away, and have the same red shift. In theory, of course this is possible. But in our actual universe, that's not what happens. In our actual universe, there is a close relationship between how far away something is and its redshift. Or, to phrase it in terms of direct observables, in theory we could have an object that was 4 billion light years away and an object that was 8 billion light years away, both with the same redshift--but they would be very different in terms of other observables like apparent luminosity and angular size. So in the general case, in theory, we would not expect any particular relationship between those variables. But in our actual universe, there is a close relationship between all three of those variables: redshift, apparent luminosity, and angular size all vary in concert and are closely correlated. That kind of observation is what leads cosmologists to conclude that our universe is expanding. > I wonder if it's pretty common to observe objects that are more blue shifted than something that is clearly closer to us. It's not only not "pretty common", it never happens except for objects very close to us, cosmologically speaking--for example, the Andromeda galaxy is somewhat blueshifted, indicating that it is moving towards us, while the Magellanic Clouds, which are closer (they are satellite galaxies of our own galaxy), are not, AFAIK, blueshifted. But all of these objects are in the same local, gravitationally bound group as our galaxy, so their relative velocities don't tell us anything useful about the large scale behavior of the universe.

elefanten 4y ago

Yes, in my 5min review it came up very quickly that “distance” and “light years” have limited uses in astrophysics. In the abstract, they reference the “z” variable of redshift in light (that has traveled for billions of years during which cosmic expansion continued and expanded the original distance) and kiloparsecs (referring to the much closer span of Milky Way through which the emissions traveled, from which they are drawing inferences about dust density therein). Wild, cool stuff!

andrewflnr 4y ago

It's perfectly possible to say when things happened at Alpha Centauri. Relativity doesn't really start messing with the notion of simultaneity until you're in a different reference frame, i.e. moving at a different speed from the observer you're comparing yourself to. But if the only difference is distance, and you know the distance, you really can just run the speed of light backward and get a meaningful time in your own past for a distant event you're just now getting the light from. Think of it this way: since you're moving the same speed, there's no time dilation between your clocks, so you can safely line up ticks between your clock and a hypothetical clock far away on Alpha Centauri or whatever. I assume things get a little more interesting at big redshifts, because you really are moving at relativistic speeds wrt what you're observing. I don't know off hand how they handle that. I will say I'm pretty sure the reason cosmologists tend to talk in redshifts is that redshifts can be pretty much directly measured and actual distance is... tricky. If you haven't already looked into the "cosmic distance ladder" you're likely to find it interesting. Source: we covered this stuff in a college class. My math is very rusty, but I did do the math.

galaxyLogic 4y ago
> if the only difference is distance, and you know the distance, But are we not moving in relation to Alpha Centauri at a fabulous speed, because all the galaxies are moving one way or the other and rotating as is Earth moving around the Sun and rotating around its axis?
- andrewflnr 4y ago
  No, Alpha Centauri and Sol are both in orbit, in the same direction, around the Milky Way. So I mean, the relative speed between us and Alpha Centauri is probably pretty big compared to a car, but it's just noise in our orbital velocities, which are still peanuts compared to c.