I managed to get through my 15-20 minute "talk," and just as I threw it open for questions Second Life had a database problem and everbody in-world had to be logged out.... We got back in 40 minutes or so later, and I answered questions for a while for people who came back. However, if you were at the talk and wanted to ask questions but didn't come back, I'll be doing a follow-up Q&A session tomorrow (Wednesday August 1) at 10AM PDT at the same location.
Below, I've got a transcript of the talk I gave. Other than fixing some typos and merging things into paragraphs, I haven't edited what I said/typed.
What I want to do is spend about 15 minutes going over these slides I
have hanging around the room. I've given 70 minute talks on this stuff
before and still had more to talk about, so obviously this isn't
everything :). What I'll try to do is explain how we actually *measure*
the expansion history of the Universe, which is the evidence that led us
to believe that the Universe is accelerating.
First, though, a few brief words about myself, and then a few brief
words on SL training. I got my PhD from Caltech in 1997, although I had
basically finished all the work in Fall 1996. I worked on infrared
spectroscopy of active galactic nuclei. It was when I went on to my
post-doc that I started working on the Universe as a whole. I worked
with Saul Perlmutter and the Supernova Cosmology Project at Lawrence
Berkeley Lab. It was an exciting time to be there, for it was in 1998
that we and our competitors both announced data that showed that the
Universe was accelerating. This has been one of the truly exciting
discoveries in Cosmology in recent decades. Not *quite* as exciting as
the data from the 1960's that confirmed the Big Bang, but it's getting
up there. This has also excited a lot of the particle physics
community, because for the Universe to accelerate, it must be filled
with something-- that we call Dark Energy-- that is not in the standard
model of particle physics.
So what I'll do now is talk about how one actually goes about measuring
the expansion of the Universe, and figuring out that the expansion is
accelerating. First, though, I want to make sure that people know how
to zoom in on the slides I have. I just highlighted one side -- the
border on the side is a lighter color. If you're sitting, you can use
the rotate left and rotate right buttons (A and D) to look around and
find it. Move your mouse over the highlighted slide, hold ALT, click
and hold the left mouse button, and you can zoom in on the slide. (By
moving the mouse.) The writing is too small to read from where you're
sitting, but if you do that you should be able to read it. Is anybody
having trouble reading the slide with the highlighted border?
Measuring distances in astronomy is really hard. There aren't tape
measures long enough to measure the distances between stars.... As
such, there are a whole slew of methods for measuring distances, and
sometimes they come with huge uncertainties. One of the most reliable
is the "method of standard candles," which this slide
outlines. Conceptually, it's very simple. Basically, you find something
whose luminosity -- the intrinsic amount of light it puts out -- is
known. Then, from how bright it is, you can figure out how far away it
On this slide, I've go two candles -- although a 100W lightbulb might
be a better example, because two 100W lightbulbs from the same
manufacturer will put out the same amount of light. The dimmer one is
farther away-- and we can quantify that. The standard candle we used
for the accelerating universe work was a Supernova.
Everybody press ESC to reset your view 😉 Hanging in the middle of the
room is a "Nova/Supernova Progenitor."
There's a big red star that's bulged out on one side. Orbiting around
the star is a white dwarf. The big red star is perhaps 100 to 1000
times the radius of the Sun. The white dwarf is no bigger than the
Earth. The gravity of the white dwarf pulls some of the outer layers
off of the red star, which goes into an "accretion disk" swirling around
the white dwarf.
As material builds up on the white dwarf, it reaches a critical mass
where it suddenly explodes and completely disrupts itself in a massive
thermonuclear explosion. Each time one of these puppies explode, they
put out pretty close to the same amount of energy. They are also, for a
few weeks, as bright as a whole galaxy. Thus, we can see them very far
away. These are the standard candles we've used.
Move on to the next slide i've highlighted.
Astronomers have a time machine-- and, indeed, our time machine is way
better than the one geologists and paleontologists have :). Because
light moves at a well-known finite speed, if you see something far away,
you are seeing it as it was in the past. However long the light took to
reach you, that's how far in the past you're seeing it.
When you go outside and look at the sun -- not a good idea, by the
way, if you don't want to go blind -- you aren't seeing it as it is
right now, you're seeing it as it was 8 minutes ago. The discovery of
the accelerating Universe was based on supernovae that exploded as much
as 8 or 9 billion years ago. Because we can measure the distance using
the method of standard candles, we can also measure exactly how far back
in time we're looking.
That's the first piece of the puzzle. The second thing we want to
measure is just how much the Universe has expanded since them time of
Move on to the next (now highlighted) slide.
Again, hold ALT, move the mouse over the slide, and hold the left mouse
button to zoom in. Many of you have probably heard of redshift --
probably because of the Doppler shift. Something that is moving away
from us will show a redshift -- light (or sound) emitted by it will be
shifted to longer wavelengths. Often, the redshift from the expansion
of the Universe is described this way, but in fact that's not really
what it is. In fact, the dynamics of the Universe as described by
Einstein's General Relativity tell us that as the Universe expands,
wavelengths of light expand at *exactly the same rate*. This is called
the "cosmological redshift".
In the diagram at the bottom of the slide, at "Time of emission" there
are two galaxies; we are the one on the right, and the one we want to
observe is on the left. At emission time, some light is emitted. Time
passes, the light travels, and the Universe expands. By the time the
light reaches us, the Universe has expanded -- so the galaxies are
farther apart -- and the light's wavelength has also expanded; longer
wavelength light is redder. The amount of the redshift we observe is a
*direct* measure of how much the Universe has expanded.
Go to the next slide, on the other side of the mollusk... 🙂 [Edit
added to transcript: The mollusk was Joshua Linden.]
The two things I've told you give us everything we need to measure the
expansion history of the Universe. Measure the distance of a standard
candle to figure out how far back in time we're looking; that's the X
coordinate of a point on the graph. Measure the redshift to figure out
how much the Universe has expanded since that time; you can use that to
figure out the relative size of the Universe at the time of emission as
compared to now. That relative size is the Y coordinate of each point.
Plot all your points for supernovae at different distances -- that is,
different lookback times -- and you have your expansion history.
At that point, we can compare it to what theory predicts. On the next
slides are the predictions from theory that we *thought* we were going
to be comparing to when the project was started in the 1990's.
Basically, mass creates gravity; all the galaxies are pulling towards
each other, which will tend to put the brakes on the expansion. If
there is a *lot* of mass, there's a lot of gravity, so the expansion
will be slowing down a lot. In that case, there may be enough mass to
stop the expansion and cause the Universe to recollapse. On the other
hand, if there isn't very much mass (the "low-mass" Universe), the
expansion should be slowing down less, and the expansion will continue
forever. However, because gravity is attractive, no matter what you
should expect the expansion to be *slowing down*-- decelerating.
Go to the last slide.
Again, hold ALT, move the mouse over the slide, push and hold the LMB,
and move the mouse. When we really made the measurement, we can up with
the result that was unexpected by most of us.... For the last 6 or 7
billion years, the expansion of the Universe has been *speeding up*.
A lot of people didn't believe this at first. One prominent theorist,
Rocky Kolb, apparently said in one talk that "the Supernova people had
better figure out what is wrong with their data, or somebody is going to
get hurt." However, there were two independent teams that came up with
the same result, so people took it seriously. Now, almost a decade
later, independent methods have confirmed that it seems that the
accelerating Universe measurement is correct.
Because gravity is attractive, normal stuff can't make this happen. For
the Universe's expansion to be accelerating, there must be *something
else* in the Universe. It turns out that Einstein's General Relativity
*does* allow for a very exotic material which will have (effectively) a
negative gravitational effect. Today we call that Dark Energy. From the
fact that the Universe's expansion is accelerating, we know that Dark
Energy makes up 2/3 to 3/4 of the total energy density of the Universe.
The big slide over the DJ's stage summarizes some of this again, and
shows a bunch of the actual supernova data that I worked on back in
1996-1999 (and may also include some additional data form a 2003 paper I
wrote -- I'm not sure).
At any rate, I'll stop there -- that's the whirlwind tour of just how
one goes about measuring the expansion history of the Universe.
For the rest of the time we have, I'd be happy to answer questions about
any of this, or about the discovery.
...and right then was when the grid crashed. I couldn't have timed it better if I tried.
If you didn't come back at about 11AM PDT to ask questions, drop by the same spot tomorrow at 10AM for a Q&A.