Monday, November 7, 2011

... The Rest of the Story Part I: Temperature

     This blog will actually contain posts about going to the Pole, I promise.  But I want to take the opportunity now when I have a little time to keep explaining the background science and why I'm actually going to the Pole.  I think it always helps to have a big picture in mind.  So thanks for your patience.

     Let's see... I left off with describing what the CMB is and why it's important.  I mentioned it has an incredibly pure blackbody spectrum, meaning the radiation has the same temperature no matter where you look out into space.  (That's a bit incorrect - a photon doesn't have a temperature, but you can infer a temperature of the matter that emitted it by looking at how the intensity of the light changes with color, i.e., its spectrum).  This fact proves a Big Bang took place.  But as with everything in science, look a little deeper and the answers aren't so black and white.  The temperature of the CMB, now a meager 2.7 K, isn't exactly 2.7 K everywhere you look.  From spot to spot on the sky the temperature changes ever so slightly. One spot, say in the direction of the Big Dipper, might be 2.7 K plus just a tiny bit, and another spot a couple full moon diameters away might be 2.7 K minus a little bit.  We call these changes temperature anisotropies.  When something is isotropic it means it looks the same from every direction.  When something is anisotropic, it looks different depending on where you're looking.

     CMB temperature anisotropies are really tiny, so tiny in fact it took us 30 years to first observe them.  The CMB was observed (accidentally) for the first time by a couple of scientists (Penzias and Wilson) at AT&T Bell Labs in 1965.  (They won Nobel prizes for this... I'm telling you CMB science is super important!  I should also point out that several theorists had been predicting the existence of the CMB for decades, so while it was accidentally discovered, the discovery didn't surprise anyone, except for maybe Fred Hoyle.)  No matter where they looked in the sky they saw this extra microwave light and it looked to be the same temperature everywhere.  For 30 years scientists tried to find deviations from this constant temperature, and only with NASA's COBE satellite in the early '90s did we finally see them.  30 years!  How small are these fluctuations?  Well, let's say the Earth's surface is as smooth as the temperature of the CMB (the Earth is still the same size, I'm just re-scaling surface topography).  If that were the case, Mount Everest (20320 feet above sea level) would be a pathetic 1.5 feet tall.  This corresponds to deviations at a few parts in 105.  That's tiny.  You've got this whomping signal at 2.7 K, and scientists are trying to measure tiny temperature fluctuations at the level of 10-6 K (10 μK, or micro-kelvin) and smaller.  That's really hard.

Penzias and Wilson in front of the Bell Labs Horn Antenna with which they observed the CMB for the first time.


     But, who cares?  Why does it matter that the temperature of the CMB isn't the same in every direction?  Nothing is perfect, right?  Well, precisely! And we owe our existence to that fact, (and gravity).  The CMB was created a long time ago, when the Universe was only 380,000 years old.  Since light can't travel faster than light speed (duh), it takes a long time to get here... 13.3 billion years.  So, when we look at the CMB, we're looking at what the Universe looked like 13.3 billion years ago - the CMB is a baby snapshot of the Universe.  This was before stars and galaxies and planets.... there was just a bunch of hot hydrogen and helium gas.  That's it.  None of the heavier elements (except for a bit of lithium) even existed.  They would be made in the cores of stars tens of millions of years later.

     Anyway, remember that the temperature of blackbody radiation tracks the temperature of the matter that emitted it.  That means a hot spot in the CMB is a region of space where matter was just a bit denser and hotter than the average.  A cold spot is a region where matter was just a bit less dense (rarefied) and cooler than the average.  Over time, gravity amplified these tiny differences.  Hot/dense spots would become even denser, and cool/rarefied spots even less dense.  Eventually, as gravity did its thing the hot spots would become galaxies and clusters of galaxies (e.g., the Coma Cluster), while the cool spots would become giant cosmic voids where next to nothing exists (like the Boötes Void).  So, it turns out that CMB temperature anisotropies are the seeds of structure in the Universe.  When we look at the CMB, we're looking at the blueprint for what the Universe looks like today.  That's powerful, and amazing.

     The power of these temperature anisotropies can be turned into quantitative predictions about the contents and evolution of the early Universe, when it was a relatively simple place before gravity mucked it up and made stars and galaxies and planets and us.  You can study how large the anisotropies are as a function of the size of a patch of the sky you're looking at.  So, let's say you observe a chunk of the sky 10 degrees by 10 degrees square.  First, you measure the deviation from the average CMB temperature for all patches that are, say, 5 degrees square within that 10x10 area.  If you were to plot these temperature up, you'd see a nice bell curve (a Gaussian) with a mean at the CMB average of 2.7 K, with some standard deviation (how wide the bell curve is).  The wider the bell curve, the larger the deviations from the mean of 2.7 K.  If you keep repeating this, and sample the 10x10 patch of sky at smaller and smaller intervals -  a degree, a tenth of a degree, hundredth, and so on -  you can plot up the size of the deviations (the width of the bell curves) as a function of the size of the patches of sky you broke the original 10x10 patch into.  We call this a power spectrum.  For the case of the CMB, you have temperature squared (the power or amplitude of the signal), on the y-axis and the angular size of  sub-patches on the x-axis.  Fitting models to the bumps and wiggles in the CMB temperature anisotropy (TT) power spectrum tells you all sorts of information about the Universe, like how much matter there is compared to dark matter, the geometric shape of the Universe, how much helium there was compared to hydrogen right after the Big Bang, and a whole lot more.  (If you're curious to see how changing parameters like this affect the shape of the CMB TT power spectrum, I highly recommend this site by NASA: the Build a Universe tool.)

     Take a look at the two images below.  The one on the top is a map of CMB temperature anisotropies.  Blue/black spots are cooler than the average, yellow/red warmer, and green are spots that show very little deviation from 2.7 K.  The image on the bottom is the TT power spectrum of this map, and shows how big the temperature fluctuations are as a function of angular size of patch on the sky.  You'll note that the power spectrum has three large bumps, and then a tail at smaller angular scales of bumps that slowly die away.  The largest bump happens at about the degree scale.  Now look back at the map on top.  Notice how your eye can see tiny clumps and patches?  The angular size of the patches your eyes are picking out is about a degree.  You're eyes are actually picking out the first peak in the CMB TT power spectrum!  (This is a bit different than what I just described, but it turns out power spectra are related to Fourier transforms, and your eyes actually take the Fourier transform of light as it passes through your lens, so your eyes are really good at taking and interpreting power spectra).

WMAP 7-year map of CMB Temperature Anisotropies.

CMB TT Power spectrum. Power (temperature [μK] squared) on the y-axis, angular size of patches on the sky on the x-axis.

     This is fantastic!  If we measure the power spectrum of temperature deviations in the CMB, we can learn about the contents and evolution of the Universe in a quantitative way.  No longer must we just say "the Universe is expanding" or "the Universe is 10-20 billion years old."  We can pin these down more accurately.  Using the CMB and a few observations of other objects and phenomena, and by fitting a model to the power spectrum bumps and wiggles, we can say the Universe is 13.76 +/- 0.11 billion years old, the Universe is spatially flat to 1% (as opposed to positively curved like a beach ball or negatively curved like a pringle), regular matter makes up 4.5%, dark matter 22.6%, and dark energy 72.9% of everything in the Universe, and lots of other more technical parameters.

     For the past 20 years scientists have been measuring temperature anisotropies with greater and greater precision, and to smaller and smaller angular scales.  But does information in the CMB stop at temperature deviations?  No way!  Turns out, we expect light from the CMB to be weakly polarized.  Polarization is the orientation of light.  Remember that light is made up of alternating electric and magnetic fields.  Polarization is defined as the direction that the electric (E) field of the light is pointing.  Mapping out how the light's polarization changes with position on the sky tells us even more information.  That's what I've been working on - a camera that's sensitive to CMB polarization to replace the temperature sensitive one on the South Pole Telescope (SPT).  But let's leave that for the topic of a future post.

No comments:

Post a Comment