DESI Data Strengthens Evidence Of Change In Dark Energy

FLORA LICHTMAN: This is Science Friday. I’m Flora Lichtman. One of the mysteries of the universe is, why does it expand at the rate it does? Back in 1998, two teams of researchers observed that not only was the universe expanding but the rate of expansion was increasing. That observation was the basis for a concept now known as dark energy. And in the years that followed, cosmologists have been trying to get a better handle on how dark energy works and where it comes from.

This week, researchers on a project called DESI, the Dark Energy Spectroscopic Instrument, released results based on their first three years of data. And when you line up their results with those from other measurements, it hints that, possibly, dark energy, whatever it is, has changed over the lifetime of the universe. The so-called cosmological constant is not a constant.

Joining me now to try to explain are my guests. Andrei Cuceu is a postdoctoral research fellow at the Lawrence Berkeley National Laboratory, and he’s part of the DESI project. And Adam Riess, Bloomberg Distinguished Professor at Johns Hopkins University and the Space Telescope Science Institute and cowinner of the Nobel Prize in 2011 for those first expansion measurements I mentioned. Welcome to you both.

ADAM RIESS: Thank you.

ANDREI CUCEU: Thank you, Flora. It’s great to be here.

FLORA LICHTMAN: Adam, in your universe, how big of a deal are these findings?

ADAM RIESS: They’re really exciting. I mean, if these hold up– and this is already the second go around where they have maintained– this would be the biggest clue we have about the nature of dark energy since it was discovered, I would say, in about 25 years.

FLORA LICHTMAN: Andrei, what did the data show? What did DESI find?

ANDREI CUCEU: Right. So in our case, we’re using one particular tracer that looks at the large-scale structure of the universe, and we’re looking at these ripples that were left frozen in this large-scale structure in the universe. And then these ripples were free to expand with the universe, which means it acts sort of like a ruler, a ruler that expands with the universe. And then we go and we measure this ruler at different periods, looking back in time. And this allows us to map the expansion history of the universe for about the last 12 billion years.

And when we take our measurements and we combine them with other results from other collaborations from other types of tracers, we start to see some very strong hints that dark energy might not be a cosmological constant, that, in fact, it might be evolving over time.

FLORA LICHTMAN: OK, Adam, big picture, a constant is not a constant. Why does that matter? What are the implications?

ADAM RIESS: Well, it sounds like we’re just talking about some number. Is that number changing or not? What we’re really talking about is our fundamental understanding of the physics of the universe and of gravity itself because if it’s a constant, if it’s what we call the cosmological constant, then the physics of that is likely that there is energy built into space and time that is sort of ever present and continues to appear as the universe grows. If, instead, as these results suggest, that it’s not constant, then we have to think about a very different picture in physics. It might be that this energy is kind of like a field but that it has been changing over time, in which case the story of the past and future of the universe is very complex. It might be that we don’t understand how gravity operates on the largest scale. That is, we may have finally broken Einstein’s theory of general relativity. And so while it sounds like we’re just talking about some number– is this number constant or is it changing– what we’re really talking about is the big picture and how we understand the universe.

FLORA LICHTMAN: Can we back up a second? How should I be thinking about dark energy? Is it a force like gravity? Is it a thing?

ANDREI CUCEU: So I think the first thing that I want to emphasize is that what we’re measuring is, essentially, the acceleration of the expansion of the universe. That’s the direct thing that we’re measuring. And then we have this dark energy that is meant to explain that acceleration. And then what this new data is indicating is that this acceleration doesn’t quite behave like we would expect it to behave if this dark energy was a constant. And this implies that there is more complex behavior going on. And this, just like Adam said, opens up a large space of possibilities that could explain this behavior.

FLORA LICHTMAN: So I want to stay with this for a minute more. Adam, when I think about dark energy, I think of it as a thing. Should I actually just be thinking of it as a placeholder for like, we’re seeing a thing that doesn’t make sense with our physics, and we’re calling that dark energy?

ADAM RIESS: Yes. The answer is both of those. So I will say the idea of dark energy, it’s not a completely made-up concept. So I’ll just take you back to high school. In physics, In Newtonian physics, you have matter. Matter attracts other matter. There’s attractive gravity. That’s nice and simple.

In Einstein’s theory of gravity, things get more complicated. We need to know about the basic physics of the kind of material. And, in particular, if there is energy in empty space, unlike matter, it can have repulsive gravity. It can push things further apart. And so that already is a pretty exotic concept, but we didn’t make up that idea.

What we did see 25 years ago and now here is that the expansion of the universe is accelerating. And so matter wouldn’t do that. So we went to Einstein’s dumpster, and we pulled out the cosmological constant. And we said, OK, well, that is a possibility for sure. And so we went with that in some ways because scientists like to shave with Occam’s razor. We like to take the simplest approach possible.

But now, as the DESI team is showing, if there is greater complexity there, if it is not just constant, then that might not be the right story. And so we’re just barely understanding the physics of dark energy. We’re just getting sort of perhaps the first wrinkles in that story.

FLORA LICHTMAN: Do these measurements get us closer to figuring out what dark energy is or get us closer to understanding what’s broken about our current understanding? Like, oh, maybe our understanding of gravity isn’t quite right, or can we drill down to any specificity about what needs revising?

ADAM RIESS: Well, yeah, we can, actually. I mean, if you take these at face value, they suggest that dark energy is weakening in some way. And, of course, a big question has always been, what is the ultimate fate of the universe if it’s accelerating and you have this dark energy? And in the past, if we thought it was a constant, it meant the universe would expand forever, in which case everything would become more and more dispersed. We would lose energy. It would be kind of a cold, empty future for the universe.

But these new results open up other possibilities, including still the possibility for a recycling universe where things recollapse in the future and we’re sort of on the n-th cycle of a continuously reforming process. So again, these results, I think, do bring us closer, at least, to getting real answers for some of these questions.

FLORA LICHTMAN: Andrei, I heard you used the word this data strongly hints. How confident is the team that this is a real effect that you’re seeing?

ANDREI CUCEU: Right, so that’s a great question. When we had our first data release last year, then we were seeing hints when we combined our data with what was available back then, and what happened now is we got significantly more data. So the precision of our measurements increased. And then what we did is, besides combining with the probes that we checked last year, we also went back to the drawing board and tried many different types of combinations with either subsets of those probes or with entirely new probes that have appeared since then. And what we’re seeing is a coherent picture where whichever combination we try sort of falls in the same region that hints at these deviations from a cosmological constant.

Now, the level of that deviation depends on the exact combination. And again, no combination gets to the level of claiming a discovery. But we’re definitely seeing a lot stronger hints this time because we’re trying many different combinations, and again, they’re all hinting at the same region.

FLORA LICHTMAN: Adam, what would you want to see to make you as confident as you could be in this result? Is it more data? Is it data from a different project?

ADAM RIESS: I would say, as an outside observer for DESI, I would say the work they’ve done is outstanding. This is absolutely state of the art instrumentation. The team that’s doing it is really the A team. This is the best people working on this problem, and this is their second iteration after hearing from the community all their ideas on the first iteration and doing very important tests. So this is certainly impressive, and it is probably more than a hint. And I think that there are many experiments coming up over the next few years that should be able to really dig into this.

I will also say that there has been sort of growing evidence that we may, indeed, be seeing cracks in what we call the standard model, this mixed dark energy, dark matter model. The expansion rate of the universe doesn’t seem to match between the early universe and the late universe, a problem known as the Hubble tension, and that has even greater statistical significance than the present result. The present result has been growing. So once you start to see more and more cracks in the model, while we don’t yet understand what’s going on, it does kind of build some level of momentum for, hey, there’s something a little more complicated here than we thought.

FLORA LICHTMAN: This seems like very humbling work.

ADAM RIESS: It is. I mean, this is the ultimate, I would say, cosmic humility, that you’re trying, as these tiny little creatures on a tiny, pale, blue dot somewhere in Nowheresville in the universe, and you’re trying to understand everything. And you can’t even go to all these places, so you look with telescopes. And like watching some complex chess game, you’re trying to figure out the rules of the game. Sometimes you think you have it. Then things surprise you.

One of the challenges, of course, is that 96% of the universe, the dark energy and dark matter, are fundamentally different than us, and so we cannot use our own intuition because most of the universe isn’t like us. It doesn’t follow our intuition. So we have to use physics, and it’s a challenging problem.

FLORA LICHTMAN: I love that idea that most of the universe is totally incomprehensible to us.

ADAM RIESS: I mean, I think it’s the most ambitious effort of humanity, really, is to try to understand the big everything. And there’s been tremendous progress in this area in the last several decades.

FLORA LICHTMAN: Andrei, talk a little bit about how you actually make these measurements. What kind of instruments are involved?

ANDREI CUCEU: What DESI is, it’s both a collaboration that actually does this experiment, but it’s actually an instrument that we mounted on this telescope in Arizona at Kitt Peak that, essentially, the main goal is we want to build a 3D map of the universe. And to do that, the most challenging part is the third dimension, which is distance between us and those galaxies that we’re trying to map. And to measure this distance, we measure it through these things called redshifts. And to do that, we have to take spectra of each of those galaxies.

So we have an army of 5,000 robots. And for each observation, each of the robots is essentially targeting a single object, a single galaxy, and collecting the light from that one galaxy. And then this 5,000 fibers that collect that light get fed into 10 spectrographs that actually measure those spectra. And then we go and we look at each of the spectra, and we measure the distance to those galaxies.

So our analysis actually takes a long time. And for the for the first-year analysis, it took us about two years from the moment when we finished collecting data to the moment where we finished all of the analysis and we’re ready to announce the results. And now with the second analysis, it took us a lot less. It took us about a year. So we finished collecting this data set in April 2024, so almost exactly one year ago.

FLORA LICHTMAN: This is Science Friday from WNYC Studios. If you’re just joining me, I’m talking with Adam Riess and Andrei Cuceu about new research into dark energy. Andrei, I heard that there was a reveal, like a big reveal of this data. Will you tell me about that?

ANDREI CUCEU: Right. Yes. So an important part of our analysis is that we blind ourselves to the result, and this is because we want to avoid human biases. Everyone has their preferred model, but that shouldn’t influence the science and the results, right?

So to achieve this, when we do the analysis, we inject something that essentially shifts the final result, which means that we see the result, but we don’t know if that’s the real result or which way it’s been shifted. So, for most of the last year, we saw this result, but it was sort of in the wrong spot, or we didn’t know where exactly it really is, but it was beautiful. It was a really precise measurement.

And then we spent most of this time digging into our pipeline, digging into our analysis, and trying to validate each step of the process and making sure that there’s no choices that were made along the analysis that have a significant impact on the result. So, essentially, we’re trying to prove that our result is robust.

And once we were happy with that process and we’re happy with all of the tests that we’ve done, in December at our collaboration meeting, we actually sat down all together, and we unblinded the result. And what that means is, for the first time, we removed that thing that we injected, and we looked for the first time at the actual result. And that was an incredible moment.

FLORA LICHTMAN: Should I be imagining people around a conference table with a velvet curtain around a computer screen? What was it like?

ANDREI CUCEU: So, yeah, it was about 200 people in a room and then one person, essentially, presenting the result. So that person, a couple of hours before that presentation, they actually ran the code to produce the results. And then all of the plots were quickly made and put in a presentation, and then, together, we take in those results as a collaboration.

FLORA LICHTMAN: Were there gasps?

ANDREI CUCEU: Yes, there were some gasps. I think the best way to describe it is the first few minutes, everyone was just sort of taking in the results. It was a lot of silence. Everyone was just staring and trying to think about the implications of what we’re seeing. But after those few minutes, there were a lot of excitement and applause, yeah.

FLORA LICHTMAN: Adam, you were one of the researchers to discover dark energy, which basically told us that we had a lot more to understand about the universe. This may be a funny question. Did that initial discovery or these sort of subsequent discoveries that add complexity to this picture, have they changed you or how you see the world?

ADAM RIESS: I think that it instilled in me, and still does, this great excitement for science, the kind of work that Andrei has described where there is great mystery out there, but we want to understand it. And we expect to learn things, but we also expect to be surprised. This is not just adding another digit to pi or something like that. These are big questions, and we have now tools to answer them through rigorous sort of accounting, hard science, experimental data.

And so, to me, it’s always made this just a really exciting endeavor. And when you see new results like this, it’s sort of like Christmas. You get to unwrap a package, and you’re like, oh my gosh. I had no idea we would receive this. And so I think it really injects the whole field with a lot of energy, not just dark energy.

FLORA LICHTMAN: Thank you both for joining me today.

ADAM RIESS: My pleasure.

ANDREI CUCEU: Thank you.

FLORA LICHTMAN: Andrei Cuceu is a postdoctoral research fellow at the Lawrence Berkeley National Laboratory, and Adam Riess, Bloomberg Distinguished Professor at Johns Hopkins University and the Space Telescope Science Institute.

Copyright © 2025 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/