Thursday, March 21, 2013

The History of the Universe (very, very abridged)

Today, the first results from the Planck mission are being released. This is an event that many of us have been waiting for since the launch of the Planck satellite in May of 2009. However, in order to explain why the Planck mission is so important, I'll first need to talk about some cosmology. Otherwise, for you non-scientists, why would you care about anisotropies in the cosmic microwave background?

Cosmology is essentially the history of the universe. Like history in general, we know pretty well what has happened recently because most of the evidence is pretty fresh and relatively easy to find and interpret. As you go back in time, however, history becomes more and more uncertain as evidence gets fuzzier and fuzzier. You can eliminate some of the crazier theories about the past because you know pretty well what happened next, but the uncertainties become greater. At that point, you can only really speculate about what happened in the past until you get some really, really good evidence. Luckily for us, we do have some really, really good evidence that tells us what the universe was like in the very early days, but I'll get back to that.

So, what are some of the fundamental cosmological observations we've made about the universe?

We'll start with the accidental discovery made by Edwin Hubble in 1927 that the universe was expanding. Hubble measured the distances to galaxies outside of our Local Group using a type of star known as Cepheid variables. Cepheid variable stars are very bright, pulsating stars whose pulsation period depends directly on their intrinsic brightness. The so-called period-luminosity relation is shown graphically below. Therefore, if you can measure the period of a Cepheid variable (which you can do over pretty large distances because they're so bright), you know how bright it really is. By comparing how bright an object really is to how bright you observe it to be, you can determine how far away it is (because you know light drops off as the inverse square of the distance).

The standard Period-Luminosity relation with period measured in log(days) and luminosity given in the funky magnitude scale, where smaller numbers mean the object is brighter. Image credit: Pietrzynski et al. 2010. Nature 468
Hubble also measured the galaxies' redshifts. What this means is that he measured the Dopper shift of particular lines in the spectrum of the galaxy. The Doppler shift you measure for an object can tell you whether or not it is moving towards or away from you. In the case of redshift, you observe that a particular spectral like that you know very well is shifted to a slightly longer wavelength (in the visible part of the spectrum, this would make it appear more red, hence "redshift"). The amount by which the line is shifted tells you how fast the object is moving.

So when Hubble combined his measured distances with his measured redshifts, he saw that all of the galaxies he observed were moving away from us. Even better still, the farther away they are, the faster they're moving. This simple observed relation is known as Hubble's Law, and directly relates the recessional velocity of a distant galaxy (far enough that it's not affected by the gravity of galaxies in the Local Group) to its distance as
v = H0d 
where v is the recessional velocity, d is the distance, and H0 is Hubble's constant. Hubble's constant is easily one of the most important parameters to measure in all of cosmology. In fact, NASA even launched a very well-known space telescope with the primary mission of measuring H0 that bears Hubble's name. As of now (before the Planck release), the most accurately known value of His 69.32 km/s/Mpc. To explain the units, for every 1 million parsecs (3.25 million lightyears) an galaxy is away from us, it is moving about 70 km/s faster.

What are the main consequences here? First off, Hubble's discovery brought the prevailing view of the steady-state universe into question (it should have thrown it completely out the window, but astronomers are often very stubborn). Second, if you know that the universe is expanding now, that means it was smaller in the past. Specifically, if you know the value of Hubble's constant, you could invert it to figure out the age of the universe. Currently, this yields an age of 14.12 billion years. This calculation of age assumes a constant expansion rate, which is certainly not the case, but we'll get to that later. If every other possible factor is taken into account, then we get an age of 13.77 billion years.

Still following from Hubble's law, if the universe really was smaller in the past, then that means that all of the matter and energy in the universe used to be more tightly packed. If it was more tightly packed, the matter and energy densities would be higher, and the universe as a whole would be hotter. If we go back in time far enough, the densities should be high enough (from everything getting squeezed together), that the universe would be both hot and opaque (everything is dense enough that radiation will always have something to interact with). Since you guys know I love to reference Kirchoff's laws of spectroscopy, you should know what's coming next: a hot opaque body produces a thermal spectrum.

This was predicted in a now-famous paper written by Ralph Alpher and George Gamow. But because Gamow, Alpher's advisor, had a very geeky sense of humor, he added Hans Bethe on the paper, such that the author list became "Alpher, Bethe, Gamow", and is now known as the alpha-beta-gamma paper (Alpher was not amused). In this paper, Alpher predicted that a hot, dense universe should have emitted thermal radiation that would be visible today, but very, very highly redshifted (increase in wavelength by a factor of ~1,100). In the same paper, Alpher also predicted that the first elements could be formed in the very, very early universe, corresponding roughly to the first 17 minutes of the universe's history. Better yet, his predictions of the abundances of light elements (mostly hydrogen, helium, and lithium with trace beryllium and  boron) were mostly spot-on.

The prediction of the highly redshifted thermal radiation from the early universe was a big deal even then, so many astronomers set out to find it. They weren't having too much luck until a pair of Bell Labs engineers (Arno Penzias and Robert Wilson) accidentally stumbled across a funny background signal in a long-range radio receiver at Bell Labs in Holmdel, New Jersey. (See! Good things can come out of New Jersey!) This was the predicted thermal radiation from the early universe. For this work, Penzias and Wilson won the Nobel Prize in Physics for 1978, and Alpher got little to no recognition for his contributions.

Since this initial observation, three different satellites have been launched to study the cosmic microwave (microwave because it has been redshifted to background. The first of these was the Cosmic Background Explorer (COBE) mission, which, for the first time, measured the spectrum of background radiation. The result was simply astounding: the universe is one of the most perfect thermal emitters ever observed. This discovery won a Nobel Prize in 2006 and has been immortalized in the following xkcd comic.

More interesting than the remarkable perfection of the cosmic microwave background was the very slight imperfections, or anisotropies in the background. Measuring these anisotropies was the goal of the Wilkinson Microwave Anisotropy Probe (WMAP). WMAP was launched in 2001, and was retired into a close orbit around the Sun in October of 2010. The most well-known result from WMAP was the following image of the anisotropies in the cosmic microwave background.

Keep in mind that these fluctuations are enhanced for the sake of the image, but are actually absolutely tiny, 1 part in 10,000.

There is a lot of very important information contained in this one image, and that is a subject that has been covered thoroughly in many papers over the years, so I won't even try to get into that here. The most basic point to know, however, is that these tiny fluctuations in the temperature of the cosmic microwave background come from very small changes in the density of matter in the early universe. The reason they are so important is because these tiny fluctuations are what eventually grew into the large-scale structures of the universe that we observe today.

The newest mission to measure the cosmic microwave background, the European Space Agency operated Planck mission, just this morning released its first results, which sparked this post so I can talk about what Planck has found while you guys still know what the heck is going on. In short, Planck is a better version of WMAP, which will work to measure these fluctuations even more precisely than ever before, which will give us even better information about the content of the universe.

The last important point to cover when providing an introduction to cosmology is the existence of both dark matter and dark energy. While many people have heard these buzzwords before, you don't often get too much of a context as to what these terms mean, and how we know they're real.

Dark matter is, as it kind of sounds like, matter that we can't see directly. We can detect dark matter indirectly by observing its gravitational influence on normal matter. Its existence was first strongly postulated by the infamous old fart Fritz Zwicky in his measurements of galactic clusters. Zwicky used the brightnesses of galaxies in a cluster to determine their masses (the brightness of the galaxy relates to the total mass of stars in the galaxy) and measured the motions of the galaxies in the cluster. From his calculations, there was no way that the clusters could be held together without a lot of unseen matter contributing its mass and gravity to the cluster. But because Zwicky was such an old fart, no one really took him seriously.

More evidence for the existence of unseen matter came from the work of Vera Rubin (who is an absolutely adorable old woman now, and I've been fortunate enough to attend one of her talks). Rubin worked on measuring the so-called "rotation curves" of galaxies. A rotation curve tells us how the rotational speeds of objects in a galaxy change based on how far away they are from the galaxy's center. This can be measured by looking for the redshift (and blueshift, which is the opposite of redshift) of spectral lines from the galaxy, but rather than looking at the galaxy as a whole, like Hubble did, one looks at different locations in the galaxy. Like Zwicky, Rubin saw that the motions of stars in the galaxies she looked at could not be explained purely by the luminous matter content of the galaxies.

There is other, completely independent,  compelling evidence for the existence of dark matter in the universe that we have only been able to find more recently. We can actually get information about the dark matter content of the from the fluctuations in the cosmic microwave background. Another very important measurement we can make is to look at how matter in our universe is grouped together. In general, galaxies exist in large clusters (like those studied by Fritz Zwicky), and the separation between galaxy clusters is very strongly dependent on the amount of total matter (normal stuff + dark matter) in the universe.

Now we're going to jump back a bit to measuring the Hubble constant and the Hubble Space Telescope. In the early 1990s, two competing groups of scientists, the Supernova Cosmology Project led by Saul Perlmutter and the High-Z Supernova Search Team lead by Brian Schmidt. (Note that astronomers and cosmologists use the letter z to denote redshift, so a "high-z" search is a high redshift search. From Hubble's law, this means they're looking at objects that are far away.) Both groups used a special type of supernova known as a Type-Ia (type one-A) supernova that occurs when a white dwarf becomes too massive (though some highly debated process that will not be discussed here) and explodes. It turns out that Type-Ia supernovae (plural of supernova because "nova" is Latin) explode in a very consistent manner, and emit roughly the same amount of energy every time another one goes off. Like Cepheid variables, when we know how bright something is intrinsically, we can determine how far away it is.

Both groups used the Hubble Space Telescope to measure Hubble's constant, and did so for the universe as it was a few billion years ago. Adam Riess, a researcher with the High-Z team analyzed the data and found something weird and totally unexpected. When he went to check with the other team doing the same work, they saw the same thing. Rather than tell you what they found right off the bat, I'll show you the graph. The top half of the plot shows the distance to the galaxy (proxied by the apparent brightness of the supernovae) plotted against the redshift of the galaxy. If Hubble's law is 100% correct, the data should plot as a straight line with a slope of H0. The bottom half shows the residuals, that is to say, the data if you subtract off the straight line predicted by Hubble's law.

So, what do you see? Does it look like the data fall on a straight line? If you said no, you came to the same conclusion as Adam Riess did. What's the big deal, you ask? If the actual trend is above the line one would expect for Hubble's Law, then you have a universe that is not only expanding, but accelerating as well.

Yeah, that's some crazy stuff. Why would our universe be accelerating though? Wouldn't the cumulative gravity of everything in the universe have the opposite effect? The solution cosmologists have come up with so far is dark energy. While we have even less of an idea of what dark energy is than we do dark energy, we know that it exerts a repulsive force counter to gravity on the scale of the entire universe such that, rather than slowing down or coasting, the rate at which the universe is expanding is steadily increasing.

The last thing I will show you here is the composition of the universe, which we have determined from studying the cosmic microwave background, how large-scale structures in the universe are distributed, and a number of other methods that I don't even understand fully, and will not attempt to explain. Do note that this is the data from before the Planck data release. Planck's results will be discussed in my next post.

I do hope you have enjoyed my whirlwind tour of cosmology. I apologize for the length, but there's just so much to cover to make sure that everyone is up to speed such that the Planck results will actually make sense. There's way more that I could also have covered, but I tried to stick to the basics and connect them through a somewhat coherent storyline as well. Thanks for sticking through to the end, and tune in tomorrow for my write-up on the Planck data once I've had time to digest it.

No comments:

Post a Comment