Last week, we began talking about understanding the size of the Universe, and we continued this week with some information on distances and motion in the Universe. This brings us to my favorite application, which leads to the Hubble expansion:
Redshift. You see, whenever an atom or molecule emits light, it gives off that light at a very few particular wavelengths. For instance, if you have hydrogen, you'll always get light at wavelengths of 656 nanometers (red), 486 nm (cyan), 434 nm (indigo), 410 nm (violet), and 397 nm (on the border of violet/ultraviolet):
Now there are three things -- and only three things (unless you really want to get technical) -- that can happen to this light to change the wavelengths that you see. Let's go over what they are.
1. Gravitational Redshift. If you're deep in a gravitational field (like close to a black hole), you have to use up energy to climb out of it. For light of all types, energy and wavelength are very closely related to each other. Smaller wavelength = higher energy and larger wavelength = lower energy, so if you need to climb out of a strong gravitational field, you lose energy, and therefore your light gets shifted towards the red. This is what we call redshift, where something happens to make the wavelength of your light longer and lower in energy. But gravitational redshift is rarely significant; two other effects are far more important.
2. Redshift due to motion. If an object that emits light moves away from you, the light from it gets redshifted. This is the same exact effect -- the doppler shift -- that causes police sirens to sound lower pitched when they move away from you. One thing that's neat? If a light-emitting object moves towards you, the light gets blue-shifted, and becomes more energetic! (We see this happening for the Andromeda galaxy, one of the only ones in the Universe that moves towards us.) And although this is incredibly useful, this is not what's happening to light in the Universe. Remember, I told you that these distant galaxies aren't moving, the space between them is just expanding. Well, guess what?
3. Expanding space causes a redshift! (And thanks to av8n.com for the image!) You see, as space expands (above), the wavelengths of the light in it also expand, as you can see below.
And this last effect is so important for the expanding Universe. Why? Well, if we measure the light from many, many distant objects and determine their distances, we can -- simply based on the objects' redshifts -- learn the entire history of how the Universe expanded. The redshift isn't hard to measure, either:
It is from literally millions and millions of these individual measurements that we've been able to determine the entire history of how the Universe expanded. That, among other things, is how we discovered dark energy and the accelerating Universe! Pretty remarkable stuff, and yet, not intuitive at all.
So what should you take away from this? That as light travels through space and space expands, it causes the wavelength of that very light to expand, too. And that's how we learn about the history of cosmic expansion in our Universe. Again, it's expansion that's causing this redshift, and not motion. Hope this helps shed some light on some of the most confusing stuff out there!
Physical Science
Comments
Ha! I finally understand Gravitational Redshift now...thanks!
Posted by: Brando | August 5, 2009 3:05 PM
... and when light interact with anything (gravity, magnetic fields, "dark matter" = placeholder for anything we don't know) it will loose energy, kind of friction in mechanics: since energy and wavelength are related, redshift will happen with or without expansion of the universe
Posted by: Bernd | August 5, 2009 7:30 PM
A very clear and easily understood explanation with outstanding graphic. You clearly have a gift for teaching.
Posted by: Art | August 5, 2009 8:54 PM
I have a question. Let's say that two objects are some distance, x, apart, and a photon leaves one and heads for the other. The distance between the objects is n times the photon's wavelength, and there is no expansion of space between the objects, so the photon travels n times it's wavelength before hitting the second object. If we repeat this process but this time with expansion, then your number 3 says that the expansion of space would stretch the photons wavelegth throughout it's journey until it hit the second object with some wavelength greater than it started with. It would also travel more distance and take more time to reach it's destination than the first photon. My question is, would the photon still have travelled n wavelengths? In other words, if n1=x/lambda in the first scenario (where lambda = the photon's wavelength = constant), would n2=int(dx/lambda) = n1 in the second scenario (where lambda = f(x))?
Posted by: Trey | August 5, 2009 9:21 PM
Trey,
Only if the expansion rate is exactly a constant over time. In that special case, then yes, the number of wavelengths traveled will be exactly the same as it would've been in a static scenario.
However, in our Universe, the expansion rate changes with time. So in your theoretical case, yes, but only because you are using a constant expansion rate. Otherwise -- if you model the real Universe -- the decreasing rate that space is expanding before the photon gets there will skew your answer towards having more wavelengths than it would have in a static Universe.
Posted by: Ethan Siegel | August 5, 2009 10:02 PM
How can you differentiate the expansion redshift from the movement redshift? How can you determine that a galaxy's redshift is caused by the expansion of the Universe, and not simply by the movement of the galaxy away from us in a "solid" Universe?
Posted by: Autofocus | August 5, 2009 11:38 PM
Autofocus,
If I remember my cosmology correctly, you can't distinguish the cause of the redshift by itself, but there are several other observations that can discriminate between a static and an expanding universe. But I will defer to our good host.
Posted by: Anthony | August 6, 2009 12:35 AM
If the Andromeda galaxy is moving towards us, then some galaxies must be moving yes? You said "I told you that these distant galaxies aren't moving, the space between them is just expanding." and it has thrown me slightly. Some Galaxies must be moving yes, acting on gravity and such? If the Andromeda galaxy is moving towards us (are we also moving towards it?), then does that mean it us heading towards us faster than the rate at which the universe is expanding?
Posted by: piratebrido | August 6, 2009 2:11 AM
piratebrido: As I understand it, distant galaxies are redshifted due to the expansion of the universe. Now, these galaxies are also moving (within their galaxy clusters), but expansion is a much larger factor. The Andromeda galaxy is within our own galaxy cluster, and is actually moving on a collision course with us (in a few billion years, mind), which is why it's blue-shifted.
Note: I'm not a cosmologist.
Posted by: Joel | August 6, 2009 2:41 AM
Theoretically.... could it be true that if I go really really fast with my car, I could reach a speed were I would see the red light of the traffic light as green as I get closer and then red again as I leave it behind?. Will the traffic lights become obsolete? Should we switch the colors?
Posted by: Funtio | August 6, 2009 5:48 AM
@Joel
That's what I had thought. Just looking to get it clarified as his statement seems to suggest that distant galaxies don't move and expansion accounts for the red shift.. If distant galaxies don't move, then why does Andromeda appear to be moving against expansion towards us; what's the difference? I could be reading too much into this, but it is a facinating subject.
Posted by: piratebrido | August 6, 2009 5:51 AM
Unless it is closer to the centre of the galaxy and is expanding out with us, thus it appears blue shifted from our perception? Like looking out the rear window of a car at the car behind us?
Posted by: piratebrido | August 6, 2009 5:53 AM
I could reach a speed were I would see the red light of the traffic light as green as I get closer and then red again as I leave it behind?
Sure.
Will the traffic lights become obsolete?
I hope to heck that we're using automated controls by the time we're going that fast. ;-)
Posted by: Naked Bunny with a Whip
| August 6, 2009 6:34 AM
Galaxies are clustered into galaxy clusters and groups, and within these they revolve around a center of gravity (just as planets do). Now, there's a lot of space between those distant galaxies and us, and it's the space itself which is expanding. So, compared to the amount of movement they're doing, they are expanding away from us quite rapidly - some are moving towards us, but they're all expanding even faster away from us. Which is why they're redshifted.
Now, Andromeda is the nearest galaxy to our own, and we're both in the same galaxy cluster. So there's not "much" space between us and them, so they're not expanding away from us as fast. However, it is moving within the local group toward us, at a rate which is faster than the rate it's expanding away. Therefore, blueshifted.
Posted by: Joel | August 6, 2009 6:55 AM
Piratebirdo,
Galaxies move with what we call "peculiar velocities", which is to say that, due to the gravity of the stuff around it, each galaxy "moves" relative to the space around it at a few hundred to a few thousand km/second.
However, the Hubble expansion rate, once you get farther than about 20-30 Mpc away, dwarfs these peculiar velocities, and is [b]independent[/b] of them. It's like having a road that slopes slightly uphill with a lot of bumps in it. At first, the bumps may confuse you, but the farther up the road you go, the more you realize it's just a bumpy, uphill road. In this case, the bumps are the "motions" and the hill is the Hubble expansion of the Universe.
Posted by: Ethan Siegel | August 6, 2009 7:31 AM
This is probably a really stupid question, but I'll ask it. If light from Andromeda is blue-shifted as it moves towards us, it's gaining energy. If, however, I was on a planet in the Andromeda galaxy and I could watch that light moving away from me, would I not see it red-shifted, and thus loosing energy? Given that light is described as both a wave and a particle, how does it have both energy levels?
Posted by: John Wilson | August 6, 2009 7:39 AM
Can you distinguish between "movement redshift" and "expansion redshift" physically, or is it the "it has to be expansion because it's faster than light" argument.
Posted by: Mu | August 6, 2009 8:03 AM
John Wilson: You couldn't see light that was moving away from you. You'd see other light, also from Andromeda, which is moving towards you.
Posted by: Joel | August 6, 2009 8:06 AM
Mu, physically, "movement" leaves other signatures besides redshift. But if all you measured was the redshift of one object, there would be no way to discern.
That's really the key to understanding cosmology, is to recognize the best way to put all the puzzle pieces together. If you're a fan of equations, then conceptually, redshift works like this:
Redshift*speed of light = Hubble Expansion + peculiar motion
Since Hubble expansion is just the Hubble Constant times the distance an object is away from us, we can get this value for a very large number of objects. Whereas peculiar motion always tops out at around 3,000 km/s, we can measure redshifts for objects with distances that are megaparsecs, 10s of megaparsecs, 100s of megaparsecs, and gigaparsecs away!
By time you get out to 100 megaparsecs or more, the peculiar motion is insignificant, and so we know we're looking at expansion. It also helps that it's the same rate everywhere in space *and* in all directions. Other explanations can be made to fit the data, but they are contrived (i.e., they place us -- and nobody else -- exactly at the center of the Universe).
Posted by: Ethan Siegel | August 6, 2009 8:50 AM
Thanks Ethan, that makes sense.
Posted by: Mu | August 6, 2009 10:11 AM
When figuring the expansion rate do you have to subtract the red shift due to the gravity of the galaxy the light is coming from? Or does the gravity well have to be black hole sized before it matters.
Posted by: the backpacker | August 6, 2009 11:42 AM
This is the best blog on earth, hands down.
Posted by: Magpie | August 6, 2009 9:02 PM
I have a question regarding the expanding universe redshift. Where can i find out more as to why an expanding universe causes or influences the expansion of wavelengths, and are there measurable effects for wavelengths above and below the visible spectrum?
Also, what books would you recommend to a beginner for gaining a theoretical grasp of physics and/or basic astrophysics (My last formal class in physics of any sort was before i started studying logistics at the university, with the exception of statistical analysis)?
Finally, another question: is it possible that an expanding universe may also cause particles to expand in any way?
Thank you for writing this blog. I am an avid (if occasionally lost) reader, and your blog is one of the few theoretical physics outlines on the net where i can read about astrophysics and "cool stuff happening in space" that i can follow without a graphing calculator in one hand and a theoretical maths textbook in another.
Please keep on writing; i assure you i will keep on reading.
Posted by: dumbassnoob | August 12, 2009 10:39 PM
Ooppss!... I've been lazy and only now I translated it into Brazilian-Portuguese here.
Sorry, Ethan!...
Posted by: João Carlos | August 14, 2009 7:04 PM
The Economist recently published this item:
Correction: In "As important as Darwin" published on August 15th, we said that no astronomer can look beyond a distance of 13.7 billion light years. This was incorrect. The universe has expanded during the 13.7 billion light years that light has been zipping across it and, as a consequence, astronomers can see to distances of perhaps as far as 47 billion light years.
I'm baffled. Isn't the universe about 13.7 billion years old? How can anyone look back 47 billion light years? Will someone please comment on the issue, which seems to fly in the face of the usual view that we can peer back some 13 billion years or so to the earliest galaxies, but no further than that.
Posted by: Colin Norman | August 29, 2009 3:57 PM
If we were in a galaxy near the edge of the universe and looking toward the edge, would we see darkness or would we see a "mirroring" of our own universe? If so, would the mirrored light be red-shifted further than if we looked directly at it?
Posted by: Steve Clay | October 17, 2009 7:36 PM
@Colin: The correction had its own error: "The universe has expanded during the 13.7 billion light years..." They mean years (a time period) not light-years (a distance).
You're correct: No photons are older than 13.7B years, but many of them have traveled farther than 13.7BLY (the distance). Expansion is like upgrading from a 14" TV to a 47" TV in the middle of a baseball pitch: The pitch doesn't take longer but the ball (photons) now has gone farther across your living room.
Posted by: Steve Clay | October 17, 2009 7:52 PM