Physics for the 21st Century logo

Section 3: Discovery of the Expanding Universe

The surprising discovery that the universe is not static resulted from a long series of observational developments in astronomy. Establishing the distances to other galaxies and their recession from us was the work of many minds and hands. Building telescopes, making instruments to record and analyze the light they gather, deducing the properties of bright stars, applying measurements of those stars to measure distances, and heroic efforts to measure the spectra of galaxies all built the foundation for the discovery of cosmic expansion.

Edwin Hubble at the telescope.

Figure 4: Edwin Hubble at the telescope.

Source: © AIP Emilio Segrè Visual Archives. More info

While Einstein was pioneering the theory of gravity, a technological revolution was under way in astronomy. Using photographs, astronomers began to measure the size of the Milky Way and began to study the fuzzy "nebulae" mixed in among the point-like images of stars. They found it difficult to determine what these pinwheel-like objects were because they did not know whether they were nearby small systems where one star was forming or distant large objects as big as the whole Milky Way. Distance measurements in astronomy are notoriously difficult and full of subtle errors. We can judge the distances to stars from their apparent brightness, but this can be deeply deceptive. If you look up on a summer night, you might see a firefly, a high-flying airplane, the planet Mars, the bright star Deneb, and M31, the Andromeda galaxy, all with about the same apparent brightness, even though it would take 1042 fireflies to emit as much light as a galaxy. To understand the real properties of these objects, we need to know how distance affects brightness.

In general, the brightness of an object falls off as the square of the distance to the object. This is called "the inverse square law." To use the inverse square law to determine the distance, however, you need to know how bright the object is to begin with. Stars come in an astonishing variety, from the dimmest brown dwarfs to the brightest blue supergiants. The power output, or luminosity, of stars whose distances we determine from geometry ranges over a factor of 1010. If we were foolish enough to assume they were all the same as the Sun, we would introduce huge errors into our picture of how the Milky Way is organized. We need to find a way to identify stars that have the same luminosity—some stellar property that doesn't depend on the star's distance from us. math icon See the math

A particularly helpful method came from studies of stars in the Magellanic Clouds, nearby satellites of our own Milky Way that we encountered in Unit 10. Careful studies of repeated photographic images revealed giant stars called Cepheid variables whose brightness increased and decreased in a rhythmic vibration repeated over a few days. Henrietta Swan Leavitt, a "computer" who studied the Cepheids, pointed out that "It is worthy of notice that... the brighter variables have the longer periods." Cepheids are what astronomers call standard candles, objects of known luminosity. If you find a Cepheid that has the same period as one of Henrietta Swan Leavitt's stars, no matter how bright it appears, you can assume it has the same luminosity. If you know its intrinsic luminosity and how bright it appears, you can infer its distance.

Large Magellanic Cloud.

Figure 5: Large Magellanic Cloud.

Source: © ESO. More info

Measuring distances

Edwin Hubble, working at the Mount Wilson Observatory, home to the world's largest telescope, conducted a campaign to find Cepheid variables in the largest nebulae, M31, M33, and NGC 6822. His goal was to determine their distances and to find out whether they were small systems in the Milky Way or distant systems as big as the Milky Way. By 1925, he had a good sample of these vibrating stars. Like Henrietta Swan Leavitt, Hubble was able to measure their periods and their brightness from photographs. For Cepheid variable stars with the same periods, hence the same luminosities, as Leavitt's stars in the Magellanic Clouds, Hubble's stars appeared about 225 times dimmer. Since the brightness depends on the square of the distance, that meant that his stars were about = 15 times more distant than the Magellanic Clouds. This placed these "nebulae" far outside the Milky Way. Einstein had tried to model a universe of stars assembled into the Milky Way, but Hubble glimpsed a much grander system. The universe was made up of galaxies as large as the whole Milky Way, separated by distances 10 times as big as their diameters.

To measure galaxy distances, astronomers use the light-year, the distance that light travels in one year. Since the speed of light is 3 x 108 meters per second and a year lasts about 3 x 107 seconds, a light year is about 1016 meters. A nearby star whose name we know, like Sirius, lies about eight light-years away; today, we see the light that left Sirius eight years ago, but the light from M31 takes about 2 million years to get here. So, the telescopes we use to view it act as no-nonsense time machines that allow us to record the earlier phases of cosmic evolution that occurred in the distant past. With modern equipment, we can easily detect galaxies billions of light-years away, and hence billions of years back in time.

The redshift and cosmic expansion

If Hubble had done nothing more than measure the true scale of the universe by measuring the period and brightness of Cepheids in the spiral nebulae, he would have sealed his place in the history of physics. However, he also used a different kind of measurement of the light from the nebulae to show that we live in a universe that is nothing like the static mathematical model that Einstein had constructed a few years earlier.

Light from the galaxies would be stretched out by cosmic expansion, much as the sound from cars zooming by on a highway stretches out as they recede. For light, this means features in the spectrum of a receding galaxy are shifted to the red. For decades, measurements of galaxy velocities determined this way were accumulated at the Lowell Observatory by Vesto Melvin Slipher.

Hubble diagram, plotting velocity vs. distance for galaxies outside our own.

Figure 6: Hubble diagram, plotting velocity vs. distance for galaxies outside our own.

Source: Recreated from original plot by Edwin Hubble, March 15, 1929. More info

In 1929, Hubble combined his distance measurements from Cepheids with those velocity measurements in the form of a graph. Today, we call this plot the Hubble diagram. While the data were crude and limited, the relation between them was unmistakable: The velocity was proportional to the distance. Nearby galaxies are moving away from us slowly (and some of the very nearest, like M31, are approaching). As you look farther away, however, you see more rapid recession. You can describe the data by writing a simple equation v = H x d, where v is the velocity, d the distance, and H, the slope of the line is called the Hubble constant. We know the equation as Hubble's Law. math icon See the math

The Hubble diagram shows a remarkable property of the universe. It isn't static as Einstein had assumed based on the small velocities of the stars in the Milky Way back in 1917; it is expanding. Those stars are not the markers that trace the universe; the galaxies are, and their velocities are not small. Even in this 1929 version of the Hubble diagram, the velocities for galaxies extend up to 1000 km/sec, much larger than the velocity of any star in the Milky Way.