Welcome to A Day in the Half Life, a podcast about the exciting and often unexpected ways that science and technology evolve over time. I'm Aliyah Kovner, a science writer at Lawrence Berkeley National Laboratory. And today we're going to talk about dark energy. Dark energy is the name given to the phenomenon that is causing the universe to expand at an accelerating rate, rather than expanding at a constant rate, or slowing in its expansion before eventually collapsing in on itself -- which are two other models for the universe that scientists used to debate over. This accelerating expansion was discovered more than 20 years ago by two Nobel Prize-winning teams. And yet we still don't know much about the how or why of dark energy. In fact, the only way to describe it currently is to describe what we know it is not. And it is not baryonic matter, which is the stuff that planets and stars and people are made of.
It is also not the same as dark matter and invisible stuff that doesn't interact with light or other forms of electromagnetic radiation. But we know it exists because of its gravitational effects on normal baryonic matter, but whatever dark energy is, it is estimated to make up about 70% of the stuff that exists. Once you factor in the equally mysterious dark matter. Our observation suggests that baryonic matter -- and remember, that's us and everything we can see -- counts for less than 5% of the universe. So if you like learning about fundamental mysteries, this episode is for you. I am so lucky to have astrophysicist Saul Perlmutter as a guest. As the leader of one of those two Nobel Prize-winning teams, he is one of the experts on the accelerating universe. I'm also excited to be speaking with Claire Poppett, an instrumentation scientist working on the Dark Energy Spectroscopic Instrument, or DESI, an instrument that seeks to study the effects of dark energy across 11 billion light years of space.
Early career researchers like Claire are making huge contributions to the development of new technology, like DESI, that will allow physicists to finally investigate and poke at the foundational theories postulated by scientists like Saul, and eventually, a new generation of researchers will be able to build improved models of the universe and better understand its origins.
Hi, ok, we've got everyone online. Thank you both so much for joining me. So the very first thing I wanted to ask you both is how you explain dark energy to someone who is not a scientist, like someone you meet at a party or a neighbor; because it's one of things that people might have heard about, but aren't sure exactly what it is.
Okay. Hi, my name is Claire Poppett. My degree is in physics and astronomy, but I'm really interested in instrumentation and how we can use advances in technology to help us answer the big questions in astronomy in terms of dark energy. We don't really know what it is. We only know it's there because we observe acceleration in the universe, which I definitely think Saul should talk about.
So, so I'm Saul Perlmutter and I'm a professor of physics at [UC] Berkeley and a senior scientist at Lawrence Berkeley National Laboratory. And I stumbled across dark energy while trying to measure what we thought was going to be the slowing of the expansion of the universe. It was, it was such a, a meaty project, and it took us years to the point that we were starting to pull data in. And once we did, we found that the data fit a universe that was speeding up in its expansion. And of course, from the scientist's point of view, that's the best thing in the world. You, you love catching the, the, the world doing something bizarre and it was doing something bizarre. It's, it's, the expansion is speeding up. And not only that, but we have don't know why it's speeding up and what we call that ignorance is dark energy.
So it seems like dark energy is sort of a placeholder or really cool sounding placeholder for something we don't yet know how to describe, but have we got any closer to describing it? Saul, how do early theories about dark energy compare to more recent ones?
It's actually been a remarkably fertile time for theorists because they are not highly constrained by data yet, so they can try out all sorts of ideas for what theories could explain an accelerating universe and what could dark energy consist of. And in fact, at one point I was estimating, or calculating, that there was basically a new journal paper published every 24 hours on this on this, from the theorists.
Um and because they're so unconstrained, if you ask any of those theorists, 'Is your theory, the right answer?' I think almost all of them would say, I don't know the ball is really in your court. And that's what Claire and I are involved with, which is to try to make measurements that might start to constrain what the range of possibilities could be. So, when you ask how do the current theories compare to where they were now what, 15, 20 years ago? What's interesting is that there's just an even a bigger variety of them. Obviously, if you write one, every 24 hours, you, you know, there's a lot to look at. But I wouldn't say that we yet have any one that is starting to you know, coalesce in our minds as, 'Ah, this one's beginning to explain everything.' At this stage, I think it's, it's really the exciting period is about to begin because the experiments are about to start giving data at a level that could constrain them. Up til now, the experiments that we've been doing have all been steps along the way, but even back in 1999, in 2000, we already knew that the level of precision that we needed was something that we weren't going to get for the next generation or so of the experiments.
Excellent. And so the measurements that you're talking about that, that the theorists need are hopefully going to come from this project DESI, right. So Claire what's kind of your role in there and what theories do you think that this could lead to?
Yeah, so Desi is the dark energy spectroscopic instrument. This is an instrument funded by the Department of Energy, so all publicly funded and you know, Saul was talking about measurement of galaxies and the average separation between them, which is all statistics. And so to have better statistics, we have to get more measurements. And this is where DESI comes in. You know, the real power is that we will observe 30 million objects over five years. And that's only possible with getting lots of observations of objects every single time we take an exposure with our spectrographs. We're building on the knowledge of lots of previous experiments to increase the number of objects that we can observe, you know, galaxies and quasi-stellar objects and all kinds of things.
So the real power of DESI, we have this large field of view. So when we look at the sky, we can see three degrees diameter. And if you don't know what that means, that means that we could fit six moons across our focal plane. This is a huge patch of sky we can look at it in any one time. And in that one patch, we put 5,000 optical fibers all positioned on the individual robots. So we get light from an entire galaxy that optical fiber that's about the size of a piece of hair.
I think it's crazy, we get an entire galaxy down that tiny little piece of glass and all those 5,000 fibers on the individual robots, we [can] position all 5,000 within about three minutes to a few micron accuracy. And then we send all of those 5,000 fibers down to 10 highly efficient spectrographs. So yeah, so DESI allows us to get all of these measurements with a really high accuracy to help reduce the statistics on what exactly the dark energy parameter is.
Yeah. One thing I wanted to ask you both is how you could possibly measure something that is invisible, because we can't really directly observe dark energy, but it sounds like you're doing it by observing many, many, many mind boggling number of visible things. But how do you do that? I mean, Saul mentioned that the technology wasn't there even just 10 years ago and that it was known that it wasn't there. So what are the technological that have made this happen?
So, in order to get more objects we have to have more fibers on the focal plane. So DESI, one of its earlier names was Big Boss which was built on a project called Boss. So Boss had 1,000 fibers on its focal plane, DESI has 5,000. And with Boss, every single fiber was hand plugged by technicians on to metal plates that were pre-drilled. So these technicians would spend an hour plugging in 1000 fibers by hand, and then they'd put the plate on the telescope and then they'd take, you know, a 15 minute observation and then they'd take it off. And then they'd replug fibers, and then they'd put it back on. So with DESI, you know, we're going from 1000 fibers hand plugged, to 5,000 fibers, all on individual robots. So that one-hour reconfiguration time goes to about three minutes.
And, and I don't know whether it's completely obvious from Claire's describing it, but she's been involved in this tour de force of technology because they had to invent these robots. Nobody had a, a technique for positioning, something so small, so densely packed with other things at this level of precision. And so they had to, to try out these technologies, they had to test them. They have to choose among, how they're going to manufacture them. Cause you need not only to make one of them to work and you need to make 5,000 of them work and they have to work reliably. And then you have to figure out a way to know whether or not they have put a fiber in the right place or not. And so they had to invent ways to observe these essentially the fibers to, to see whether they were going where it's supposed to go. So it was just a remarkable kind of project. In fact, it's the sort of thing that I think Lawrence Berkeley Laboratory has been is so good at, at being able to pull together these teams that really understand the science, but they also understand the engineering at the same time. But I, I just want to highlight that because she she may take it for granted by now, cause she's been doing it for so, so well for so long. [Laughing]
Haha, I 100% realize how lucky I am to be working on this project! Um but yeah, you know, this project, there's teams from all over the world all making different parts the, you know, the robots that hold the fibers are, they were designed in Berkeley, but they were made in Michigan. The prime focus corrector -- these six huge one meter diameter lenses -- they were built in different places assembled at University College, London. We commissioned a plane to bring them across to the U.S. So they could be assembled at our Fermi Lab. We've got teams from Australia, teams from Spain, teams from China, all over the U.S., and all of these groups all work together at the same time to bring this instrument together and it's on the telescope. And, you know, those fibers actually are going to galaxies, they're getting light; not right now, but they will be again in a few weeks. It's amazing.
Yeah. So I did hear that there was an initial test and that everything's working and looking great and everyone's really excited. So have the official data collecting experiments begun?
So, right before we had to close down because of COVID, that was kind of right at the end of our commissioning period, which actually was quite lucky that that's when it happened. You know, we spent four months commissioning where we the instrument was installed, but we had to figure out how it works. We had to figure out how to get those fibers on targets, how to make the spectrographs work how all of those different things were going to work. And I actually looked this number up: during that commissioning, we obtained spectra, we obtained 12 million spectra, just in those four months. Like, that's an insane number to get whilst we're figuring out how the instrument works -- 12 million spectra of 700,000 unique objects going up to high Redshift.
So what is, pardon me, I don't know these terms. Is a spectra, like one light image of an object?
So someone's described it to me before as um, turning star light into rainbows, you know, we take the star light, we turn it into rainbows, and from these unique signatures in the rainbow [we can see] like the oxygen doublet or any number of lines. You can see exactly how old that object is from that shift in wavelengths.
Wow. That makes it sound even more magical than it already did. That's amazing. So how do these huge astronomical datasets tell you about the invisible dark energy? Where are those clues in all of that information?
So there are actually only a few handles we have on this mystery. And so we're trying to use everything we we've got to get at them. Essentially the starting point for dark energy is that we know that in the earliest part of the, of the expansion of the universe it looks like the universe really was slowing down, because gravity was attracting everything to everything else. But about halfway through the expansion of the universe -- so since, you know, coming up to today -- things got far enough apart that the gravity was not as strong anymore. The distances between the galaxies were just too big. And so what was lying underneath became important, and we believe that the dark energy was lying in wait, waiting for its moment, and then it had a chance to start accelerating the expansion after the gravity you know, became weaker.
But that means that we have the possibility of being able to understand what the dark energy is, if we can trace very, very carefully this history of the expansion of the universe, going from that period in which it was slowing down and seeing how it switched to starting to speed up. And that's one of the key approaches that we're all using to get at what dark energy really is. Now, the way that we discovered this acceleration in the first place was using a rather simple technique which is treating exploding stars, supernova, of a certain kind as what we call a standard candle that we know in a fair amount of detail, how bright each one is when it's this particular kind of supernova. And so if you can see them at different distances across the universe they're fainter if they're further away. And that means that you can find out a bit about different times back in history, because they vary -- the ones that are further away are for the back in time that you're seeing, because it took time for the light to reach us from that distance explosion.
The other thing we need, then, is to know how much the universe expanded since that explosion, since we use its brightness to tell us when it occurred. Now, we want to know how much has the universe expanded since it occurred. And that turns out to be actually fairly straightforward, because you can use the fact that the light, as it travels to us, its wavelength gets stretched just as the universe gets stretched. So by the time it reaches us light that had a short wavelength -- was blue when it exploded, when the supernova exploded -- by the time it reaches us, it's been stretched to looking red. And so you can just read off from how red the spectrum becomes, should tell you how much the universe expanded since that time. So that's a fairly straightforward way of making this measurement, which is one of the few that we have to get a dark energy, but there's another approach.
That is the approach that was invented and tested in the time since we discovered this acceleration and that's this approach called Baryon acoustic oscillations. And that's the one that DESI is designed around. Now, DESI can measure, as you've heard, measure spectra of millions and millions of, of objects at a time. And in fact since the supernova project needs spectra as well, um we're hoping to use DESI for a little bit of the supernova work too, but its fundamental goal is to make this measurement of what we call Baryon acoustic oscillation using the spectra. And so, so that's, that's really what it's aimed at. So why don't I pass the ball back to Claire to describe how that works?
Yeah. So Baryon acoustic oscillations, I think is difficult to say it it's difficult to understand [laughing]. So one of the ways that I've used to describe it in the past is using an analogy as a frozen ripple on a pond. So if you throw a pebble into a pond and then you freeze the pond and observe it 13.8 billion years later, you could learn a lot of things about the pond, right? You could learn, like how deep it was or the size of the pebble that you threw or the composition of the water. And so in this analogy, the pond is the primordial plasma of the early universe. The pebble is the fluctuations and the density of the visible baryonic matter or the normal matter of the universe and then galaxies predominantly form on these ripples. So by looking at the average separation galaxies as a function of time, which is the distance between the ripples and how that distance changes or stretches, you can learn things about dark energy.
Wow, that's very cool.
I should probably, probably mention that these two techniques are very nicely complimentary in that one is working from the very earliest times that we can measure of the, the time of the cosmic microwave background glow that we see, and working your way forward in time using, you know, it's distance standard to tell you, you know, where galaxies should be in distance from each other and then how much they are actually measured to be tells you how much has expanded. And the supernova works its way the other way. We use relatively nearby supernova that are relatively recent and they tell us how much a, how bright the supernova should be, and then how bright it appears to be tells us how far backward looking and we're measuring the expansion going back from today. So the two overlap in the middle and it allows us to actually do a very strong check of the all of these techniques by seeing that we're getting the same high precision measurement in the part where they cross over.
And and so it's, it's actually a very interesting, you know, fortuitous circumstance that we have two ways to do this same kind of measurement coming from different angles.
And they agree.
Yep. Yep. And there's one other thing that we can possibly get to learn about dark energy, so far that we know of, which is that it's possible that it's not a new energy in the universe, but that it's a difference of how gravity behaves. And so one of the things that we've been, uh, that main projects are starting to do, is to look at how matter clumps in the universe and that that clumping could be different depending on how gravity behaves. So there's several other projects that have been coming up in the works, which will be coming in probably after DESI which will have a chance to try using some of those other techniques, including something called weak lensing. So I just thought it was worth mentioning that we don't have very many arrows in our quiver, but those are the ones we've got to get at dark energy today.
That's interesting. So, because not, there's definitely no set definition of what dark energy is. It could just be a property of gravity and that wouldn't even really necessarily negate previous theories? It would just fit with them. Is that possible?
I, I think that's right, but I, I think maybe the the, the interesting thing about something like gravity is that, gravity already went through one revision when we went from Newton's theory of gravity to Einstein's theory of gravity. And and, but what's interesting is that Einstein's theory of gravity had to be able to explain everything that Newton's theory of gravity explained. So the the way that philosophers of science sometimes talk about it is that every new theory has to swallow the previous theory. So if we come up with a new revision to our theory of gravity, it will have to manage to do everything that Einstein's theory of gravity does already, which is pretty tall order. Cause it's a, it's an amazingly elegant and yet amazingly bizarre theory. And so so to, you know, if that's that turns out to be the explanation, I think, well we'll be thrilled in any case, we'll be thrilled, you know, if it turns out to be a new energy that dominates the universe, or if it turns out to be a change in, in the, in Einstein's theory of gravity it, you know, both of these are just really exciting for the point of view of, of you know, the physicists.
Great. so I have a question that might be kind of a stupid question, but if you're using a standard candle in the supernova measurements, I mean, if we're not really sure exactly how gravity works or exactly what dark energy is, how can anyone really know with certainty what the light was like originally? I mean, how do you know what might've happened to shift it on its way here if we don't know what happened in the interim? Like how do we have a standard candle?
Ha ha. So, so I think the, the starting question there is how do we recognize a single event out there in that we see from, you know, millions and billions of years back in time? How do we recognize that it's the same as some event that we've seen relatively recently? And, relatively close?
And this is the magic of, of using the the spectrum in other ways, converting the bit of light that we're getting through these gigantic telescopes into a rainbow of light. So you spread it, essentially with a prism, or its equivalent. And and you look at this rainbow of light that you get. But you find is that some colors are much brighter than other colors in that rainbow. And some of them are much darker. They have both dark bands and bright shimmers at different colors in that rainbow. And those are fingerprints of the activity that went on in the star, in that case, that made up the galaxies that Claire's looking at, or in the exploding star, the supernova that we're looking at. And these are very very detailed, characteristic bits of information. So for example, when a star explodes its first, the outer layers are sending you a spectrum, and they, and you can read off in that spectrum, you know, what are the main elements?
Because you can recognize the signatures of different elements. You can read what velocity they're traveling at and when the, in the, in this explosion, because you can see how much the Doppler shift moves, the the, the wavelengths. You can read off their abundances by how intense the light is. And that is just the outer layer. Then an exploding star that becomes transparent as it becomes, as it expands out. And you start seeing deeper and deeper and deeper in. And so each of these spectra is essentially telling you about a different depth into this expanding bowl of gas, till by the end, you have a, if you watch it all the way as it expands and you get to see the entire time sequence of spectra, all these little rainbows for that one exploding, uh, star you have a CAT scan essentially of the entire event. And that's amazing. It means that we can do a CAT scan of something that occurred, some of the most distant ones we have now, about 10 billion years in the past. And in principle, if you have this CAT scan of these spectra, you can just match it up to which one does it look like among nearby CAT scans of, of, of supernova. And those are the ones you you treat as a standard candle. You know, the peers that look just like each other today and 10 billion years ago.
Okay. That's neat. And so I guess even if the light gets stretched out, the like the ratios would stay the same.
Yeah. So what you do is you look for all the same patterns, but you expect them now. Instead of being, you know, this, this pattern here in the blue part of the, of the rainbow, they've all now stretched out and they're stretched further into the red part of the rainbow. And in fact, some of them are now moving into the infrared part. And so we're needing to design new telescopes, new instruments that can pick up those spectra out there in the infrared. And that's one of the things we've been working on in our group over the, over the past years.
Awesome. Thank you. So one thing I wanted to talk about is surprises in this field, and I think it's fair to say that the field of dark energy began with a surprising research development, but in addition to that momentous discovery, what has been the most exhilarating head-scratching surprises for you both?
So I talked a little bit about DESI during the commissioning before, but I think my biggest surprise so far, you know, only been working on this project for the past 10 years is just how well DESI is working. You know, this international team have brought this instrument together. The focal plane alone has more than half a million individual parts, just on the focal plane. You know, it took three years to build and test, and after we installed it, you know, only four months later we had 12 million spectra, that was kind of huge.
So it didn't take a lot of rebuilding and trial and error, but just kind of worked.
Yeah. So, you know, we fixed a lot of things before we sent it to the telescope. You know, those three years building, if we had to build it again with all of the knowledge that we gained throughout, maybe it would only take two years, but yeah, once we installed it, it was amazing and really fun actually.
Yeah, no, I, I think this, this story of of the possibility of making really complicated, difficult things work because of this, essentially there's, it's sort of this organizational can-do spirit that these that has been developed in certain parts of the sciences that makes it possible for teams from all around the world to make an object that you never would think would be possible. And I'd say that those have been some of the most amazing surprises to be watching in these last years. And, and, you know, DESI is just the latest example. You know, obviously you know, this is the period in which we, we heard about the discovery of gravitational waves you know, as well, and that was done by, you know, teams of people doing something that if you just looked at what it would take to make this happen, you would assume that it was impossible to do.
And for me, I think it's it's almost one of the lessons of science that is least well understood by the general public, which is that humans are capable of doing very difficult things together, and that we shouldn't feel daunted by the challenges of our world. We should take them as this is something that we're great at, and that we, and if we are good at joining forces to do them, especially these international forces that we see working on a project like DESI you know, I would just would not be as, as worried by all these different things that people are scared of in the world and that, you know, I would be scared of as well. And that, and so in that sense, my, my biggest concern is that we don't know how to use that super power that we have, that we know well, how to bring people together and, and and, and feel like we're just on the same page and we're doing it together. And if we got that part working as well as the physicists have done for these projects I think it would just be a much more comfortable world to live in because we would be excited by these challenges that we faced. And we'd be working on them together.
Well, maybe we should put scientists in charge.
[Laughing] Not a bad idea and take, take special note from the astrophysicists. So, I mean, it's interesting to me because it's, it's one of those fields where the idea of of the lone genius up on the mountain at the telescope, you know, trapped in their own head with their own amazing thoughts, making these discoveries like that's long since passed. And maybe that sad to some people, because it's not as you know, it's not as romantic or as amazing of a daydream to have it. That could be your future as a scientist. But I think that this kind of project will help people show that there is there's the beauty and the romance of the huge team of the computer scientists and the roboticists and the theorists and, and all that. So, I mean, that's a really great story of this, you know, for me to take away.
Yeah, I'd never I'd never been to a telescope site before I started working on DESI. And it's absolutely one of the things that I've enjoyed the most since being in this group. Again, I looked this number up during the commissioning, which was four months long, we had 71 members of the DESI collaboration visit Kitt peak, which is the telescope site where DESI is, it's a four meter telescope on Kitt Peak in Arizona. And they were scientists, they were engineers, they were students who had all these different specialties. They was theorists and instrumentation and every single range of that you could possibly think of all coming together in this one little control room, working on DESI. And sometimes going outside and looking at the stars. In quiet times. It was amazing.
Right. So I think what we may have given up a little bit in terms of the, you know, the lone scientist hero picture we've gained tremendously in terms of the that sense of, of, of a social engagement, you know, together. And I, you know, I've sometimes been commenting to people when they're trying to solve, you know, what do they want to do with their lives you know, to students that if you if you want to do something, you know, by yourself in quiet seclusion you know, maybe you should become a writer or, or or something, you know, one of these other fields, but if you want to be in the thick of people and and, you know, be in remarkable social world where you're talking to people and bouncing ideas off of other people and working together, Come be a scientist because that's where, you know, where all the socializing is happening. You know, it's a bit of an exaggeration, I'm sure that there's a lot of socializing going on in, you know, half a dozen other fields that that you can think of that are not science, but but it isn't remarkable how it's become such a social activity now.
Aliyah & Claire (31:03):
No, it's, it's fun
So I would just like to close by asking you both kind of, we talked about the evolution of this field and where it's going next, but you know, a little bit about the personal evolution of, for both of you to get to where you are today. So what made you both want to pursue your careers that have now met around this topic of dark energy?
So I had no idea that being a scientist was a real job. I thought that I thought that if you did science at school, then you were a science teacher. But I did always love to tinker with things. And I remember from a really young age, I would take things apart to figure out how they work and then try to put them back together again. And then I went to university to study physics purely because I was good at it. And then, you know, by being in a lab environment, I learned that I could continue to tinker professionally. And by applying that to astronomy, I could make instruments that could help answer some really fundamental questions that humans have been asking for thousands of years. And you know, it's really a privilege to be able to do something that you love every day. Well, not every day, but most of the time.
Yeah. And for me, I think I I always knew that I was, I was really interested in the fundamentals of what makes things tick, you know, how does the world work? I, in fact, I originally I remember feeling like, you know, it was a little bizarre that here we live in a world and nobody gave us the owner's manual of how to work it. And, you know, isn't everybody else talking about that, you know, shouldn't, we know what to do with this thing, but I've always loved the idea that you might be able to learn something very deep about how things work, how the world works, how the mind works. And I, I also just feel very lucky that I just came upon the time when you could actually ask some of these questions. And I love that mix of very deep philosophical questions with very practical you know, organizational things where you actually have to build something and make something work and fly it from point a to point B and get these people to talk to each other. And and I think it's just, you know, it is very lucky to be able to work on something that you, that you feel gets at what you enjoy.
And everybody you work with also enjoys it that much. I think that's a huge part of this day to day working life.
Exactly. That's and that's. And I think that everybody could have that experience. I might, I think that would, that would be the world we want to live in.
Yeah. Nicely put, well, thank you both so much again for being here. It was a pleasure to talk to you, both you too. It was fun to talk to both of you. Great. Thank you so much. Okay. Bye.