Brian Schmidt: How dark energy was discovered

In this edition of Titans of Science, Nobel laureate Brian Schmidt discusses how his work led to the discovery of our universe expanding at an accelerated rate, and how dark energy was theorised to be responsible...

Chris - Why were there those two different numbers if people were just measuring objects moving away from us, seeing how far the universe appeared to be stretched in the interim, why did that give two values that were quite disparate?

Brian - Turns out it's really, really hard to measure distances in space. You can't just put a ruler down. And so we had photographic plates. Photographs are good for looking at things, but they're terrible for measuring things accurately. We had not developed precision methods for measuring distances. We had Cepheid variable stars, which were by far in a way the best thing going back to Hubble, but they only are bright enough to be seen in the very nearest parts of the universe where the universe really isn't expanding because there's too much gravity around us and everything else. So it really took the development of technology to allow us to start measuring the numbers accurately. And so that happened right at the time I did my PhD thesis. I was born at the right time. I had access to the first digital data on stars and I used the brightest stars there are, exploding stars called supernovae, to measure the distances. So Type 1A supernovae are these giant thermonuclear firecrackers and they are very well behaved. It turns out no matter how you blow them up, they kind of do the same thing. They're just firecrackers. It's like a piece of TNT and it doesn't matter if you light it with a fuse or you kick it with a hammer. When it blows up, it's still a big piece of TNT. So they are remarkably uniform compared to most things in the sky. And it turns out that just by measuring how long they take to explode, that tells you to within 5% in measuring distance, how bright they are.

Chris - Were you looking for the light or were you looking for other things that come out from supernovae things like the x-rays that they blast out. What was your signal that you were looking for when you were doing this work to spot them and then record the distance?

Brian - So this has changed quite dramatically. In 1990, we had amateurs. There was a reverend in Australia who very famously could memorise the nighttime sky and just point his telescope and say 'there's a new star there.' And he would phone them in and we would look at them and sure enough he was, he was right. Then we got digital cameras, new technology. We went away from photographs, which were a real pain to go and find exploding stars because you had to go develop it and weeks pass. So these digital cameras allowed us to use computers and start scanning all the nearby galaxies. So we were out using optical light looking for new stars that appeared around nearby and progressively more distant galaxies.

Chris - And when you say the appearance of new stars, these are stars that appear because they are blowing themselves up and becoming very bright.

Brian - That's correct. So the star was there, it just was a billion times fainter. And so when you look at it, you couldn't see anything but just black and then suddenly it's as bright as a billion stars and it really shines at that point.

Chris - This gives you a wealth of data so you can go well beyond Hubble. When did it become apparent what the real number was then? Where, between those two extremes that were born out of the inaccuracies of the recording measures that they were using decades ago, what the real number for the expansion of the universe was?

Brian - So in 1993 for my thesis, I measured a number that was almost directly in between. The number was 73, which gave an age of the universe of about 14 billion years. It made everyone unhappy. The people who got the high number thought I was wrong, the people who got the low number thought I was wrong. And I would say I was not absolutely convinced I had nailed it, but at least it was a good try. The thing that really broke, I think, the deadlock was the Hubble Space Telescope technology that allowed us to go through and go out an order of magnitude further. That allowed us to calibrate a whole bunch of supernovae using them. And then it became very clear that the number it turns out was around 70. So I got I think a gold star, and that number 70 has been refined over time. Now with the James Webb Space Telescope, the Hubble for many years. Adam Riess, my colleague, has really spent 20 years doing this. And he gets 73 and a half plus or minus a very small number, which in itself is very interesting.

Chris - The implications of that number are that not only is the universe getting bigger, but as it gets bigger, it gets bigger, faster. It's blowing up. How did that become apparent?

Brian - In 1993, I had done that work to measure how fast the universe is expanding now. That technology that allowed us to start measuring the Hubble constant accurately, these digital cameras, suddenly allowed us to start looking at supernovae, not just millions of years in the past, but billions upon billions of years in the past. So suddenly what became possible using Type 1A supernovae was to measure how fast the universe was expanding nearby, like the last 10, 50 million years. And then compare it to objects that were 10 billion light years in some cases. And so we were able to then see what the universe was doing and was it slowing down a little bit or slowing down a lot or not slowing down at all? That's what we thought, you know, were the possibilities in 1993.

Chris - What did you think? Did you think that it was just a continuous rate of growth then, or were you prepared to accept that actually we could have been wrong? It's gone through various phases over its lifetime. What was your thinking?

Brian - So the theory at the time was pretty straightforward. The heavier the universe, the more it slows down over time. And so we were going to measure how much the universe slowed down and the more the universe slowed down over the time that we looked at it, the heavier the universe was. And so we're essentially just going to weigh the universe and by weighing the universe we could actually figure out was the universe going to slow down so much that in the future it would stop, go in reverse, and we would have, as I always like to refer via Douglas Adams, the 'Gnab Gib'. The Big bang in reverse. Or was it going to expand forever and not slow down enough? So it just kept on going for eternity.

Chris - Did it kind of concern you then that when you began to get the numbers in it didn't seem to fit the 'big bang in reverse' type model? Did you think we're getting something wrong or did you think actually this is really exciting, it's not going to go into reverse?

Brian - Our first object that we got, we got really good data on it and it was showing that the universe was expanding slower in the past and sped up. And that bothered me because that was not a reasonable answer, but it was one object and we're like, 'okay, we're going to get more objects, see what happens'. And at the end of 1997, Adam Riess emailed me a figure of the preliminary results on where we were up to that time. And it very clearly showed that this was not an anomaly, but all the objects seemed to be showing that the universe was expanding slower in the past and had sped up. So that was not me yelling eureka immediately, quite the opposite. It was, 'alright, we have screwed up something we need to go through systematically and do everything from scratch with me doing the stuff you did and you doing the stuff I did' and making sure everyone, everything had a separate completely independent channel on it. And so that took a few months as you might imagine, but after that couple months, the numbers didn't move around at all. And so we had to live with it by the beginning of 1998. You know, we weren't going to make this go away.