We’re talking, once again, about common questions and misconceptions about the expanding universe. If you haven’t already done so, take a look at my posts on the fact that time isn’t expanding and the fact the there’s no center to the expansion.
Today I’d like to talk about redshift. In the 1920’s, Edwin Hubble noticed that the more distant a galaxy was from our own, the more the spectral lines of that galaxy were redshifted. All this means is that rather than measure a spectral feature at, say, a wavelength of 600nm, you might measure it at 620nm. Within a particular galaxy, every single line was shifted by the same fractional amount:
with called the “redshift”, is the wavelength that the light should have under laboratory conditions, and is the difference in observed wavelength from the laboratory wavelength. Hubble found that for relatively nearby galaxies, redshift and distance are just linearly proportional to one another. Further away, the relationship gets a bit more complicated mathematically, in part because there are so many possible types of distance.
Under normal conditions, we usually think of redshifts as being caused by the relative motion of the body emitting the photon and the body absorbing it: a Doppler shift. This, incidentally, is how radar detectors work in speed traps, or, using sound waves, why you hear a shift in pitch when a firetruck passes by. One of the most common ideas, and one that even shows up in some introductory physics classes, is that the light from distant galaxies is caused by a Doppler shift. It’s not*.
What’s really going on is that as a photon travels from the source to our telescopes on earth, the universe expands underneath it. If the universe doubles in size, so does the wavelength of the photon.
Here’s the balloon analogy to help makes things clearer:
In other words, what redshift really tells us is not how “fast” a galaxy is moving away from us, but rather, how big the universe is now compared to how big it was when the light was emitted. This picture has a number of implications:
- Photons lose energy as they travel over cosmological distances. For light, long wavelength means low energy. Consider the light coming to us from the cosmic microwave background. It originates (around 380,000 years after the big bang) at a temperature of about 3000K, comparable to the surface of a relatively cool star. After subsequent expansion of the universe by a factor of 1200 or so, the temperature is now only about 2.7K.
Since photons lose energy as the universe expands, but massive particles don’t lose mass, at some point in the past, there was more energy in the form of photons than in ordinary and dark matter combined. This was back at a redshift of z=3900 — when the universe was about 1/3900th the size it is now.
- A redshift doesn’t tell you anything about the expansion right now. For a distant galaxy, you only learn about the relative size of the universe when the photon was emitted to today. The universe could be accelerating or decelerating or even stopped at this moment, and you don’t get that directly form the redshift. What you need are lots of redshifts in order to figure out the history of the expansion, and thus the rate of change.
- If light travels through a bunch of hydrogen clouds on the way from a distant galaxy to us, the galaxy and each of the clouds will each be at a different redshift. The galaxy will be most redshifted (since it is most distant), and so on. The clouds will then absorb only the particular wavelengths of light that correspond to spectral lines. However, since for each cloud there’s a different redshift, this creates a “forest” of absorption:
I sometimes get followup questions about how we know that the redshift really does correspond to cosmological expansion.
That’s a fair question. Here is a partial list of things which are 100% consistent with the view of cosmological redshifts, but would have to be utterly discarded if it were wrong:
- General relativity (since GR predicts a dynamically expanding or contracting universe)
- Gravitational lensing
- Distance estimates from nearby galaxies
- Any and all interpretation of the Cosmic Microwave Background
- The Big Bang model
But let’s suppose you’re one of those people who never liked the Big Bang model anyway and think you’re being subversive by challenging it. (You’re not, by the way.) Before throwing out a model that matches theory and observation perfectly, you’re required to come up with a model that does at least as well.
Good luck with that.
* Well, technically it is, but only a little bit. Throughout this series of posts, I’ve been ignoring something called “peculiar motion.” Galaxies really do move around their local environment, but at a speed of only a few hundred kilometers per second. The peculiar motion causes a Doppler shift on top of the cosmological redshift.