How Do We Know What We Know?

musings on scientific knowledge

Monthly Archives: April 2011

Hubble, nebulae, galaxies, and our place in the universe

One hundred years ago, most people thought that the galaxy and the universe were the same thing.

It’s weird to think about, looking at photos like this one of galaxy M101 from the Astronomy Picture of the Day. But there are a couple of things to realize. One is that most of us have been trained to see these images as galaxies–it seems perfectly obvious, because we’ve seen lots of them with captions about being this or that galaxy and because we already accept the notion that there are many galaxies in the universe. If you think that it’s very unlikely there are other galaxies–if you are skeptical of things being so far away, when we’re already a trillion times as far away from the center of the galaxy as we are from the Sun, when the Sun itself is the about same distance as forty thousand trips between New York and Los Angeles–then maybe you look at an image like this and see shreds of gas being pulled into a newly-formed star system. But using only images of what were then called nebulae, Edwin Hubble managed to show that some of them were in fact other galaxies.

Hubble, for whom the telescope was named, presented some evidence in his 1917 Ph.D. thesis on faint nebulae that supported the nebulae he observed being outside our own galaxy, not part of it. First of all, he saw most of his nebulae in directions that suggested they were above or below the disk of the galaxy, when you’d expect to see them concentrated in the disk itself like the stars are. He says, “Suppose them to be extra-sidereal and perhaps we see clusters of galaxies; suppose them within our system, their nature becomes a mystery.” (p. 7) He also attempted to measure rotational speeds and relate them to densities and masses, and under a number of assumptions (some quite bizarre to modern eyes) found they made more sense if you assumed the nebulae were very far away, about a million light years. So Hubble was clearly on the side of the nebulae being galaxies.

There was a pretty famous debate in 1920 between Harlow Shapley and Heber Curtis, arguing about the size of the Milky Way and the existence or not of other galaxies. This was settled in 1923, when Hubble got the first distances to other galaxies, a process that’s worth talking about in itself.

One of the main limits in cosmology is that you can’t go out and measure distance directly for objects that are far away. All we can get is the characteristics of the light coming from these objects: the amount of light (the flux), the color (spectrum) of the light, changes in both those things with time. Hubble used both the amount of light and the change of light with time to figure out the distances to some of his faint nebulae.

We know the relationship between flux and luminosity. Flux is how much energy from light hits an area of a given size in a given amount of time: for example, a 1 square meter solar panel just outside the atmosphere of the Earth pointed directly at the Sun is hit by about 1400 joules of energy every second, so we’d say the Sun’s brightness is 1400 Joules/second/meter2, or 1400 Watts/meter2. Luminosity is the total amount of energy an object gives off in a given amount of time: in the Sun’s case, about 4×1026 Watts. Flux goes down as distance from the bright object goes up. If you’ve ever pointed a flashlight at a wall and walked forwards or backwards, you’ve probably got a sense of this. When you’re close to the wall, the spot is small and bright, so there’s a lot of flux; when you’re farther away, the light spreads out and the spot is bigger but dimmer. The total amount of light coming off the wall is the same, of course, because it’s all the light coming out of the flashlight, which doesn’t change; in this analogy this total amount of light is the luminosity of the flashlight. Galaxies give off light in all directions, and we can’t see all of it–so it’s like we’re a tiny speck on the wall, seeing only whatever light from the flashlight hits us, which will lessen as the flashlight moves away. The light from the flashlight spreads out on the wall, and the total amount of it we’re getting (the flux times the area) is basically the size of our speck divided by the size of the giant flashlight-spot on the wall, times the luminosity of the flashlight itself. You can describe the whole system with three quantities: the luminosity of the light bulb, the flux on our speck on the wall, and the distance between them.1

Here’s the problem, then: how do you know how bright the flashlight is to start out with? You definitely don’t know the distance, so you have to know the other two quantities in that equation. The answer is that there are a few things we know the intrinsic brightness of, and the best one is Cepheid variables, a kind of pulsating star2. There’s a relationship between the time it takes to do one pulse and the luminosity of the star, which was worked out by Henrietta Swan Leavitt and which is a topic I’m going to leave for another post. Anyway, Hubble managed to find some Cepheid variables in these distant galaxies, and from them derive the distances–vastly farther than even the largest estimates of the size of the Milky Way. This evidence was accepted very quickly by the astronomical community: there were many galaxies in the universe, not just our own, and the universe was far larger than we had realized.

1: I’m ignoring that the wall is flat and the flashlight close–the relationship we use in astronomy usually relies on the diameter of the area of observation being much less than the distance between the observer and the source being observed. We assume that we’re well into the small angle approximation regime, in other words. Hard to work that into the analogy without getting bogged down, though. return

2: Not a pulsar, that’s something different. return