Michelle Thaller
Cosmic Distances and the Virtue of Stupid Questions
Recently we had one of my favorite events at the Jet Propulsion Laboratory -- the annual public Open House. One weekend a year, the gates of JPL are open, and everyone is invited to come in and have a look around (if they ever manage to find a parking space, that is.) This year, we got 60,000 people.
I spent most of the weekend sitting in a booth with a big "Ask a Scientist" sign over the top, wearing a silly hat to seem less intimidating. I also had some nice, colorful stickers to hand out that said "I Asked a Scientist a Question!" to further encourage people to come up and talk to me. In the end, I didn't need the extra ploys. Even I was surprised at the number of people who lined up to talk to me about any science question that had been rattling around in their heads, from what makes a black hole, to where NASA is hiding the alien bodies.
The best questions are always the ones that start with a person saying "I know this is a really stupid question, but ... " Those are always the kickers. Take one question I got that day that began this way. The question was "How do astronomers know how far away the stars and galaxies are?"
Seems simple enough. We're constantly told about how huge the universe is, how unimaginably distant the stars are. So how do we know that? As soon as I heard the question I began to smile, because a simple question like that can illustrate just how clever we humans are, yet how little we really know about the universe. And I like to make people appreciate just how much we don't know, at least not just yet.
It often surprises people to learn that astronomers don't have one simple way to determine exactly how far away an object in space is. Some methods produce very accurate distance measurements, but work only on objects that are relatively near to us in space.
The way we measure the distance to the closest stars, for example, is called parallax. This method actually is straight-forward, and you can model it for yourself using your thumb. Make a "thumbs-up"sign with your hand and hold it out at arm's length (and keep it still). Close one eye and notice what objects in the background, say on the other side of the room, your thumb seems to be lined up with. Now open your other eye and close the one you had open. Notice how your thumb appears to have moved with respect to the background? That's parallax.
The reason your thumb changes position is that your eyes are separated by a small distance, allowing you to view your thumb from slightly different angles against the background. Now bring your thumb closer (maybe a foot away from your face) and do the same thing. The apparent change in position against the background gets bigger. The closer something is, the greater this parallax gets.
Astronomers use the exact same method to find the distances to the stars. Instead of viewing an object with our eyes, we widen the distance to the size of the Earth's orbit around the Sun. By taking pictures of the same stars six months apart (or, viewed from opposite sides of our orbit around the Sun), we can measure how much the positions of nearer stars shift with regard to farther away, "background" stars.
These apparent shifts allow us to measure the distances to the nearest stars very accurately, which is quite a feat given just how far away even the nearest stars are. The closest star to us, Alpha Centauri Proxima, is a little under four light years away (a light year being about 6,000,000,000,000 miles). Even with the nearest star, I find it impossible to visualize how distant that number of miles really is. Here's a fun comparison, though. In reality, our Sun is about 100,000 miles across. If the Sun were the size of the dot on an "i" on this page, how far away would Alpha Centauri Proxima be? The answer is an astounding 50,000 miles, more distant than any place on the surface of the Earth.
The problem with parallax is that it only works if you can actually measure a star's shift in position against farther away stars. With the help of some recent satellites, we've used parallax to measure the distances to tens of thousands of stars, mapping out our neighborhood in the Milky Way galaxy. But how do we measure the distances to other galaxies? Now it gets tricky.
It's probably best to switch terms at this point. When we get out of our own galaxy, instead of measuring distances, it's more correct to say we estimate them. That's not to say our estimates aren't very good, but there's just no way to know for sure whether we're off by a few hundred meters, or a few trillion miles.
Take the nearest large spiral galaxy to us, the Andromeda galaxy. With large telescopes, we can resolve the elegant spiral arms of Andromeda into billions of tiny individual stars. Of course, those stars aren't really tiny, they're just very far away. In fact, shouldn't stars in Andromeda be pretty much the same size and brightness of stars in our own galaxy? That's the reasoning astronomers use to estimate the distance to the nearest galaxies. Any source of light seems brighter when you're close to it, and dimmer when you move farther away. By measuring how bright a star in Andromeda is and comparing it to nearby stars like the sun, we can estimate how far away our neighbor galaxy is.
The problem is that stars are not all the same brightness. In our own galaxy, we know of stars that are thousands of times brighter than our sun, and many that are far dimmer. How can we estimate the distance to the stars in Andromeda, if we don't know how bright they are to begin with? The challenge to astronomers was to find a kind of star that we knew the intrinsic brightness of. And there is such a beast too.
Cepheid Variables are dying, bloated stars that, as their name suggests, vary in brightness. They pulse, growing brighter and dimmer, as unstable layers of gas in their atmospheres expand and contract. A while ago some astronomers figured out that the longer a Cepheid Variable takes to pulse, the brighter it is. When Cepheids were discovered in the Andromeda Galaxy, we could measure their pulsations and estimate the brightness of the stars, and hence their distance from us. Using that method, we estimated the distance to Andromeda to be about two million light-years. We don't know the exact distance (Cepheids might be subtly different in another galaxy, skewing our estimate), but we've got a good ballpark figure.
But what about the next step? What about the myriad of tiny, distant galaxies that are too far away for our telescopes to resolve individual stars? Out here, it takes an elegant union of observation and theory to probe the cosmic distances. For the most distant galaxies, astronomers use the Cosmological Redshift to pin down distances. The implications of the technique are quite mind-blowing.
Basically, astronomers try to measure how long the light from a galaxy has been traveling through the expanding universe. The longer it's traveled, the more stretched out it's gotten as the universe expands. It may seem weird to think of light being stretched out, but it really happens. Light is a wave, and as the universe expands, the distance from one "peak" of the wave to another peak increases. Practically, this means the color of the light changes. There's really no difference between a red wave of light (called a photon) and a blue photon that's been stretched out. Thus, the longer a photon has been traveling through space, the redder it appears.
Astronomers can measure this effect with extreme accuracy. hydrogen, the most common gas in the universe, emits light in very specific colors. When we see the chemical signature of Hydrogen in the light of faraway galaxies, we can measure exactly how much the colors have shifted. The problem comes when you try to interpret this result into a distance. The exact distance depends on how old you think the universe is, and how fast you think it's expanded in the past, things we just don't know.
We have some good guesses, but there will always be a range of uncertainty. The farthest objects we've measured "redshifts" for are estimated to be about 12 billion light years away. We may be off by hundreds of millions of light-years, but we have no further tests for accuracy. And in the end, even millions of light-years is only a tiny percent of the actual distance of these galaxies. Our big picture is still, to a large extent, secure.
The question of how astronomers know how far away things are turns out to be far from simple, and definitely not stupid. So much of astronomy, and indeed all of science, is like this. Scientists seem so authoritative when they announce the discovery of a distant galaxy, or a black hole a billion times the mass of the sun, but all these measurements have errors and tons of assumptions rolled up inside of them.
To me, science seems to be a lot more about what we don't know than what we do. And isn't that the point? We wouldn't even have started down the road to understanding the universe if someone hadn't been bold enough to ask the stupid questions.