SAO Guest Contribution




Hello, my name is Brian Schmidt and in the next few pages I will attempt to explain the latest results on supernovae, and how we use them to measure the Universe. I am an astronomer at the Australian National Universities' Research School of Astronomy and Astrophysics. We are located at Mount Stromlo Observatory, just outside of Canberra. The observatory has telescopes at two sites, one at Mount Stromlo, and the other near Coonabarabran. I highly recommend coming for a short visit to see the sights if you are in the area.

The work described here is not just my work, but the collective effort of more than 20 people from around the globe with whom I work. As the group's leader, I tend to get more than my fair share of the credit, but a project like this really represents the blood and sweat of many people. So if you are interested, have a browse, and learn about the most powerful explosions in the Cosmos, and the ultimate fate of the Universe.

 Some of the introductory material has been covered else where in this course, but you can click on the Items highlighted in red to cover that material again. 


Measuring the Universe: A Brief History

In the 19th century, astronomers and philosophers battled about the extent of the Universe. Emanuel Kant argued that we lived in a galaxy like the faint nebulae seen through telescope, and he was right, a half a century before it became generally accepted.

 The first good look at galaxies was taken by Vesto Slipher who measured their velocities using the Doppler shift on spectral lines in the object's spectra. He found that almost every galaxy was moving away from us! Want to know more about spectra and the Doppler shift?
Slipher's represented a cosmic conundrum for astronomers of the day: Since the time of Copernicus, astronomy has presumed that we are not a special place in the Universe. But Slipher's results seemingly contradicted this belief - we were a special place, the most unpopular place in the Universe from which all other objects were trying to move away. Slipher's results remained a mystery until Edwin Hubble came along in the 1920s with the then world's most powerful telescope, the recently completed 100inch telescope on Mt. Wilson, near Los Angeles. He used the physical law that an object becomes fainter as its distance increases to gauge the distances to Slipher's galaxies.

In 1929 Hubble announced his results. He assumed that the brightest stars he could see in a galaxy were all the same brightness, and found that the faster an object was moving away, the fainter its brightest stars were, thereby showing that the more distant an object, the faster it was moving away from us. He announced that the Universe was expanding. Want to know more about how Hubble's observations indicate we are in an Expanding Universe?

 We name the rate that the Universe is expanding after Hubble - The Hubble Constant. The Hubble constant tells us how fast an object is moving away from us, given its distance. If you think about it, the Hubble constant therefore tells us given an object's distance, how far apart those two objects will be at any time. If we extrapolate to the time when these two objects were on top of each other, we reach the big bang. So the Hubble constant tells us how old the Universe is. 

But here we have assumed the Universe hasn't changed in its rate of expansion over time. 

The Changing Speed of the Universe

As the Universe expands, gravity pulls on the Universe, and slows the expansion down over time. As we look to great distances, we are looking back in time. If we can measure how fast the Universe is expanding in the past, and compare it to how fast it is expanding now, we can see the total gravitational effect of all matter in the Universe.

Here we plot the distance between two galaxies as a function of time. Looking back into the past we see that the galaxies get closer together until they are ontop of each other - this is the time of the Big Bang. If the Universe expands at the same rate, it will follow the dotted yellow path. But if it is slowing down over time the Universe is younger than we would otherwise think, speeding up, then it is older.

If there is lots of material, the Universe will be expanding much faster in the past --- it will have slowed down a lot --- so much so that the Universe will eventually halt in its expansion, start to contract, and eventually end in the gnaB giB (that is the Big Bang backwards). In most models of the universe, this type of Universe curves onto itself (like a sphere), and is finite.

If there isn't much material, the Universe will be expanding about the same speed in the past as now, and will continue to expand forever. This type of universe curves away from itself (like a saddle), and therefore is without end, now, in the future, and even at the time of the Big Bang.

Here we once again plot the distance between two galaxies as a function of time. Looking into the future we see that the galaxies get further and further apart, except if gravity is able to halt the expansion. If the Universe expands at the same rate, it will follow the dotted yellow path. But if it is slowing down over time the Universe it eventually turns around and starts to contract. If the Universe is speeding up, it will continue to do so at an ever increasing rate.

A favourite model amongst theorists is for the Universe to be precariously balanced between being finite and infinite. This balanced Universe is known as a critical universe. Space neither curves away nor onto itself, it is flat, and is, for most theorists infinite. If the Universe is made up of normal gravitating matter, it will slow down in its expansion over time, however, if the Universe has some other forms of material in it, it can actually accelerate over time. 

2D representations of the shape of the Universe. Universes with lots of material curve onto themselves like a sphere. Universes with little matter curve away from themselves like a saddle. And Universes with just the right amount of material are flat.

In practice, we use Einstein's equations of General Relativity to understand what we see in the Universe. In addition to assuming his theory is right (it sure seems to be everywhere we have be able to measure so far), we do have to make a few assumptions. The most important of these are that the universe is homogenous (that is, the material in the Universe is, on average, evenly spread through out the Universe) and isotropic (matter, the expansion, and everything else is the same in all directions that we look). With these assumptions we can predict how bright an object will be given its rate of recession (the simple relation found by Hubble breaks down at large distances). 

So, if we can measure distances, we can see how these compare to the predictions of General Relativity, and in this way we can see what is in the Universe, and gauge how this material affects the Universe. It turns out this also allows us to predict what the future holds for the Universe. 

Using Type Ia Supernovae to Measure Distances

There are two major types of exploding stars which astronomers label "Supernovae". Massive stars that explode when they run out of nuclear fuel, and white dwarf stars that are ignited into gigantic explosions when they acquire material from a neighbouring star, and exceed 1.4 times the mass of the sun. These Type Ia supernovae are the key to measuring the distant Universe. Want to know more about supernovae?
SN 1994D observed with the Hubble Space Telescope. The SN is the bright star in the lower left corner. 

Type Ia supernovae in the nearby universe are observed to all have a similar brightness, and this makes them very powerful objects for measuring distances. By simply observing how bright they are, you can measure how far away they must be - the further they are the fainter they appear. In addition, because they are so bright, they can be seen at great distances, and these two things make them currently unique objects for measuring the vast distances of the Universe. Unfortunately, they are very rare. The last one seen in our galaxy was in 1006, and it must have been incredibly bright - easily visible in the daytime. 

The idea to measure the Universe with Supernovae is not new, it has long been contemplated, but it is only in the past decade that it has become feasible. The first distant SN Ia was discovered in 1988 by a Danish team, but it wasn't until 1994 that they were discovered in large numbers. Since 1995 two teams have been discovering these objects: A team which I lead, the High-Z SN Search, and a team lead by Saul Perlmutter known as the Supernova Cosmology Project. 

To measure the fate of the Universe, we need both distant and nearby objects, as it is only through the comparison of nearby and distant objects that the Universe's behaviour uncovered. Amazingly enough, the first good nearby sample was only completed in 1996 by a group at Cerro Tololo Inter-American Observatory (CTIO), and these objects are what enable us to use supernovae to measure the ultimate fate of the Universe. 

In actuality, type Ia Supernovae are not all exactly the same brightness. They vary by as much as a factor of two. But Mark Phillips, Mario Hamuy and collaborators at Cerro Tololo Inter- American Observatory in Chile showed that faint supernovae rise and fall very quickly, whereas bright supernova brighten and fade much more slowly, By looking at how much the objects faded in the first 15 days following maximum light, their work showed that type Ia supernovae can give distances which are good to about 7% - equal to the best of astronomical distance indicators. 

Discovering the Accelerating Universe

In 1995 my team discovered our first distant Type Ia supernova, SN 1995K. It was surprisingly faint - so faint that it seemed the Universe was not being slowed down much by gravity. But it was not until 1998 that we became convinced something was truly amok with our assumption that the Universe was filled only with gravitating matter. Want to know more about discovering distant supernovae?

 In 1998, our team published the figure below.

This plot shows our supernova data. Across the Bottom axis is Distance (or equivalently time since the SN exploded). The side axis shows change in the Expansion rate from its current rate.

 As you can see the points do not show the Universe slowing down enough to have a gnaB giB. Furthermore, they do not even seem to show the Universe coasting, rather they show that the Universe is accelerating! Every point lies in the bluish green accelerating part of the diagram. Simultaneously, the Berkeley Group, led by Saul Perlmutter, published data from which they came to the same conclusion!

Dark Energy

So what could make the Universe accelerate? It turns out one possibility is a form of Dark Energy invented by Einstein in the 1920s to make sure his equations of General Relativity did not predict the Expansion of the Universe. Although Einstein was willing to predict several things with his equations of General Relativity, he obviously thought an expanding Universe was too much, and so he invented the Cosmological Constant, a form of Dark Energy that repels itself, and was meant to balance the force of gravity, and make the Universe static. He later called this his greatest blunder once it was discovered, in 1929, that the Universe was expanding

 But starting in the 1980s, the Cosmological Constant has made resurgence as a cosmic panacea. Although the Cosmological constant seems capable of curing many of the ills of the modern cosmology such as the age problem: (A Universe with a lot of cosmological constant is older than we might otherwise think, therefore making the age from the Hubble Constant and the age of the oldest stars in agreement). 

Here we show how much cosmological constant and normal matter (that is the stuff you and I and everything around us that we see is made out of) there is in the Universe. The darker the color, the more likely a particular combination of the two types of matter is. The numbers are plotted as the greek letter Omega which is the symbol we use to represent the fraction of the Universe made out of matter relative to the amount necessary to make space flat. (i.e., if Omega = 1 then there is just enough matter to make the Universe flat). As you can see, it appears that the total amount of normal matter in the Universe is quite small - Omega is less than 1 for normal matter, and the amount of cosmological constant is reasonably large - Omega greater than 0 for the cosmological constant.

Supernovae are not the only way to make such a diagram. Another way is to look at the Universe when it was very young, and glowed like the sun. This radiation is the furthest thing we can see in the Universe, and is known as the Cosmic Microwave Background. By observing how the lumps and bumps in this background are distributed, it is possible to measure the amount of matter in the Universe. New measurements from this method give their own measurement of Matter and Cosmological Constant. 

If we combine the two measurement into one diagram, we can figure out what both methods together are telling us right now. As you can see the best values (the hashed red and white area where the two methods overlap) are having normal matter 0.2 of the critical value, and the cosmological constant about 0.8 of the critical value. 

This is actually quite amazing because many theorists want the sum of Omegas in all types of matter to be one - this is the prediction of a theory called Inflation - and these results suggest that it just might be. 

As mentioned earlier, we are one of two groups working on this project, and we have arrived, using different data, to almost identical conclusions at the same time, without knowledge of each other's results. The above diagrams show both teams data (they are the Supernova Cosmology Project). But just because we get the same answer does not mean we are both right, we could both be being fooled by our supernovae in the same way. Or maybe our whole cosmological model is flawed. 

So these observations may really mean one of three things: 

  • The Exciting: the Universe is accelerating. The Universe is accelerated by some unknown type of matter that is spread throughout the Cosmos. 

  • The Heretical: General Relativity is as sacred as anything in Physics, but it may be wrong. Since our work is comparing the predictions of General Relativity with observations, if General Relativity is wrong, so are our conclusions.

  • The Mundane (at least from our point of view): We are simply wrong and have been fooled by Supernovae into believing the Universe is accelerating. Maybe supernovae are fainter in the past, and therefore appear further away then they really are.



    We hope and believe it is the first alternative, but we have to work hard and test to see if it isn't the second or third alternative. The accelerating Universe is a major revolution in our understanding of the Universe around us. While the evidence is now strongly in favour of this conclusion, extraordinary claims require extraordinary proof. 

    Dark Energy Versus Gravity

    Despite other methods, Supernovae are currently the only viable method to actually measure the Universe is accelerating. But Einstein's version of Dark Energy makes a strong prediction, it should have been much weaker in Universe's past.

    Like two wrestler's vying for domination of the Universe, Dark Energy and Matter were then poised against each other in a fight for the future of the Universe 7 Billion years ago. Dark Energy pushes the Universe apart, while Matter pulls it together. By looking back even further, instead of to 5 Billion years, to 10 Billion years in the past, it is possible to see this battle when gravity was much stronger than it is currently. If Einstein's Dark Energy is the cause of the acceleration, it should have overwhelmed Gravity 7 Billion years ago - before this time, Gravity was winning the war, and slowing the Universe down. 

    Unfortunately, supernovae are very faint at distances beyond 7 Billion light years. In 1995 the deepest map of the sky was made by the Hubble Space Telescope called the Hubble Deep field. In 1997, this field was reobserved by two astronomers, Gilliland and Phillips, and a very faint supernova was discovered, called 1997ff (See figure)

    The Hubble Deep Field showing the thousands of galaxies observed in 1995. In 1997, SN 1997ff, exploded in one of these many galaxies (left bottom panel), revealed by subtracting the 1995 data from the 1997 data (right panel)

    By itself, this discovery image was useless. How far was the object, how old was the object, how fast was the object moving away? were all questions that needed to be answered. But this object was too faint to be observed with anything other than the Hubble Space Telescope. Furthermore, the observation, rather than being made in optical light, needed to be made in the infrared, because the doppler shift, at great distances, shift the light we used to study the supernova so much, that this light ends up stretching outside the visible region. Late last year it was realised by Adam Riess, a colleague of mine who works at the Space Telescope Science Institute, that a number of infrared observations were made of a tiny region of sky near the position of SN 1997ff, for a completely different purpose, just at the time this object had exploded. This serendipitous data set was just what we needed to measure the distance to this object.

    For six months we studied this data set to sift how many photons had arrived from SN 1997ff at each colour. These data showed that SN 1997ff was at a redshift of z=1.7 (meaning its light had been shifted by 170%), and was more than 10 billion light years in distance. These observations allowed us to measure how fast the Universe was expanding 10 billion years ago, and this most distant exploding star provided the remarkable answer. As predicted, the Universe was slowing down 10 billion years ago, just as Einstein's equations predict. 

    The Future

    As the Universe accelerates, galaxies are getting further apart faster and faster. Assuming this continues, the inescapable conclusion is that the expansion of space between us and galaxies we see now will eventually exceed the ability of light to reach us. Eventually, in the hundreds of Billions of years in the future, this will be true for the entire Universe outside of the nearest few galaxies. This stretching of space beyond the speed of light is not true for places where there is currently enough gravity to cause the Universe to collapse - here the Dark Energy can never take hold a stretch the Universe to infinity. So our galaxy will be spared, but we will look out onto a cold, dark and empty Universe from our vantage point in the Milky Way.

     But this is bleak prediction of the Universe's future is predicated on the idea that the acceleration will continue (which is true if Einstein's Cosmological Constant is the culprit). But other theories exist with names such as quintessence, x-matter, and Vamps. These theories allow the Universe to almost anything in the future, but can be tested to a reasonable degree, by very careful measurement of the Universe's past. Is the rate of acceleration exactly as predicted by Einstein's Cosmological Constant, or is it different. Supernovae maybe able to provide an answer - provided the difference is not too subtle - but resolution may require a whole new class of experiments that astronomers are just beginning to dream up. Only time will tell. 

    Brian Schmidt

    © Swinburne Copyright and disclaimer information
    Maintained by: Glen Mackie (
    Authorised by: Prof. Karl Glazebrook (
    Monday, 19-Nov-2007 11:17:07 AEDT

    Back to the Guest Contributions Index