Into the Impossible With Brian Keating - NEW Dark Energy Data Surprises Scientists (ft. Kyle Dawson)

Episode Date: June 18, 2025

Is everything we thought we knew about the universe… wrong? Is the Lambda-CDM model nearing its breaking point? Could dark energy actually be evolving? In this episode of Into the Impossible, I’m joined by Kyle Dawson and Daniel Green to discuss the latest data from the DESI experiment. These new results are making headlines, and rightfully so, as they hold the potential to transform our understanding of the universe completely.  Kyle Dawson, a key figure in the DESI project and professor at the University of Utah, explains the findings from DESI’s second data release. Together with Daniel Green from the University of California, San Diego, we dig into how fresh observations of dark energy, baryon acoustic oscillations (BAO), and cosmic expansion are revealing cracks in the standard model. We also break down the implications of these results and talk about the possible existence of negative neutrino masses—a topic that’s been gaining a lot of attention in the cosmology community. Don’t miss out!  — Please join my mailing list here 👉 https://briankeating.com/list to win a meteorite 💥 — Key Takeaways: 00:00 Intro 01:09 Most surprising results of the DESI experiment  02:38 DESI’s instrumentation and data analysis  05:43 What’s lacking in the CMB? 07:17 The 4.2 sigma tension in cosmological data  10:44 Theoretical models and dark energy evolution 22:16 Neutrino mass, dark energy, and paradigm shifts  28:41 Scepticism about current cosmological models 30:22 Negative neutrino mass 40:35 Technological advancements and the future of DESI 48:15 Outro — Additional resources:  ➡️ Follow me on your fav platforms: ✖️ Twitter:⁠⁠⁠⁠⁠⁠⁠⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/DrBrianKeating⁠⁠⁠⁠⁠⁠⁠⁠  🔔 YouTube:⁠⁠⁠⁠⁠⁠⁠⁠ https://www.youtube.com/DrBrianKeating?sub_confirmation=1⁠⁠⁠⁠⁠⁠⁠⁠  📝 Join my mailing list:⁠⁠⁠⁠⁠⁠⁠⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://briankeating.com/list⁠⁠⁠⁠⁠⁠⁠⁠  ✍️ Check out my blog:⁠⁠⁠⁠⁠⁠⁠⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://briankeating.com/cosmic-musings/⁠⁠⁠⁠⁠⁠⁠⁠  🎙️ Follow my podcast:⁠⁠⁠⁠⁠⁠⁠⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://briankeating.com/podcast⁠⁠⁠⁠⁠⁠⁠⁠  — Into the Impossible with Brian Keating is a podcast dedicated to all those who want to explore the universe within and beyond the known. Make sure to follow so you never miss an episode! Learn more about your ad choices. Visit megaphone.fm/adchoices

It's peak pollination season, and my business is scaling fast. To keep the nectar flowing, I need a phone plan with top priority data speed. That's why I chose GoogleFi Wireless. My connections stay strong even when the hive is buzzing. Plus, unlimited plans start at $35 a month. Now, that's a deal that doesn't stay. Explore Google Fi Wireless plans today. Plus taxes and government fees.

GoogleFi Wireless is not subject to data traffic deprioritization during times of high network usage. You said this place was steps from the water. We just haven't found the steps yet. How much did we save? Enough. Enough to get lost. Or you could book a stay with Hilton. Welcome to your oceanfront room.

Just steps from the water. The Hilton sale is on now. Book on Hilton.com or the Hilton app and save up to 20% to get the stay you expected. When you want savings, not surprises. It matters where you stay. Hilton for the stay. What I really see this is more philosophically is kind of a null test against the LAM-M

model, the standard model we've had in place for 25 years. And it's starting to see evidence that there's something wrong with that model. And there's some transition, some perturbation, some extra component that we need to add to the standard model. He took it very hard that, you know, he worked on this study of galaxies, and he would have said they should have been the ones to discover it. That it was like the fact that they had this thing staring them in the face and they let the supernova people make the discovery.

It was like, well, how are we ever going to discover anything if we had this giant signal that we were staring at? We couldn't see it. New data from DESE, the dark energy spectroscopic instrument probing our universe, revealed something surprising about dark energy that has major implications for both the future of our universe and our present understanding of fundamental physics. Today, I'm speaking with Dr. Kyle Dawson,

professor of physics and astronomy at the University of Utah, the past spokesperson of the Desi, and my colleague, Professor Daniel Green at UCSD to unpack these new mysterious findings. Let's go. Kyle, welcome to the Into the Impossible podcast. So nice to have you. Great. Thanks, Brian. The headline, of course, is that Desi has seen evolving signs of dark energy, that Einstein or good old friend, he might have been wrong. It might not be a cosmological constant.

It might be some variable constant. What was the most surprising or interesting result in data release two of the Desi experiment? For me, the most appealing thing of data release two was the fact that we were able to do a better job explaining where the information was coming from from the cosmology, we now had enough signal in the BEO measurements themselves to really see how they aligned with the LAMDA CEDA predictions on the Hubble diagram and how that differed from, say, the supernova at low right shift. What is a baryon acoustic oscillation?

BARN acoustic oscillation, or also known as the BIO, that's the acronym we use, are features left over from the very early universe. These features appear in the distribution of galaxies in any other matter that's out there. What happens is as the period of inflation ends, the fluctuations in the matter density introduce these sound waves. They produce these sound waves in the plasma and the barion fluid that exists. The barons coupled to the photons in that very early time. And these sound waves travel outwards.

They travel for about roughly 300,000 years. When the plasma evaporates or decouples, the sound waves are free. frozen for the rest of time. And what that gives you is an overdensity at a very preferred scale in the distribution of matter. It's there for all of time. And galaxies and quasars and neutral hydrogen, they all preferentially align with that overdense. All right, now we understand how BAO works. Let's understand how DESE works. So, Kyle, can you explain how DESE actually does when it does? Desi is the dark energy spectroscopic instrument. It's a spectroscopic instrument. It's instrumented on the four-meter male telescope in Arizona, Kip Peak Natural Observatory,

and the goal is to take spectroscopy of as many objects as possible at a time, with the idea of building up a very, very large three-dimensional map of the cosmos. We observe galaxies and quasars out from right-shift zero to roughly speaking three and a half or four in Redshift, about 5,000 objects at a time. How is it instrumented? Is it what technology is a single-spectro graph? It's using fiber fed. Break it down and keep in mind the audience is the most intelligent in the non-multivirce. So the instrument itself is a multifiber spectrograph. So we have 5,000 fiber optics that sit in the focal plane of the telescope.

They're robotically controlled to sync up with the location of a specific target, whether it be a quasar, a star, or a galaxy. And then the fibers feed 10 different spectrographs, each of which disperses light over a range of roughly 4,000 to 10,000 to 10,000. thousand angstroms. And from that wavelength coverage and from the spectra that we obtained with the instrument, we're able to obtain red shifts. And that's the key measurement that comes out of the data that gives us the 3D map. Professor Green, as a theorist, what excites you the most about these results? Is it the grand scope of what they could possibly reveal? Or is it the cracks that let the light come in as Leonard Cohen. Well, I think I was excited about DESE for long before they had data because it was supposed to be the sort of final piece in the puzzle for a lot of questions

we had that if you just planned out the future, Desi was going to reach the sensitivity where we were going to, you know, one of the big things we were planning for was to detect massive neutrinos. And we knew that Desi was the data set we needed to sort of piece everything together. So it's always been on the radar. It's like that's sort of one of the key pieces of the puzzle to sort of making everything fit. The cracks are unexpected and exciting, but I was going to be excited if there were no cracks, and it just all fit together perfectly, because we knew it was going to reach that point where we were finally able to really see details that we expected to be there, but yet are, you know,

exciting signs of physics that is not tested otherwise. The most exciting thing about LHC is if it sees nothing. No, no, that's a disaster. It's probably the end of your career, you know, if it sees nothing. I would say that people used to, used to, you know, I would be advising postdocs. or graduate students work for me for like a job interview, and they would often be preparing for the question, well, what happens if future cosmological experiments don't see anything?

I'd be like, we didn't see neutrinos. Oh my gosh, that's a disaster. You know, we're predicting we're going to see it. That means there's no neutrino mass. This is a massive change to particle physics. And I often would say that because it puts that question on its head, right? Because people want to be like, future's going to be so boring.

But that was why I was excited because Desi was going to be this like guaranteed signal. We didn't know exactly what the data would reveal. Oftentimes you can ask, you know, what do you hope it sees? So, as you know, Dan, in the Keating household, the bread is buttered by the CMB. Can you explain what's lacking in the CMB? I thought the CMB was the oldest light in the universe. Can it do everything from tell us the, you know, the Deuterium abundance to tell us the length of all standard rulers that one could ever care about? What are these guys even doing here? There's a lot we can learn about the universe at the epoch where hydrogen was formed, which is 380,000 years after the Big Bang, as you know, but for the listeners. But things that affect the universe later, like, say, in the recent few billion years, those are more precisely measured by things that are measuring the universe locally.

So, Supernova have been measuring the expansion rate sort of right around now. and DESE is measuring it sort of, again, in the past, you know, many billion years, but things like the amount of matter in the universe, that was really one of the key things that DESE can do better for this neutrino mass measurement, but in general, just anything that's relevant in the late universe is important just for making everything fit, right? So the CME makes certain, is able to make some of those measurements, but not at the precision of DESE. And so DESE was going to come in and sort of be that new anchor for the value of, say, the matter in the universe, that that was sort of one of the critical measurements.

And of course, the named dark energy spectroscopic instrument tells you that learning about dark energy and other things to do with expansion has obviously done better than with the CME. Kyle, in the beginning are the first data released ER1. You had some evidence for the same phenomena that we were talking about today. So there's consistency between the data sets, which is important. Are there, you know, kind of gold standard, you know, objects within the data set from all the different things you mentioned, QSOs, galaxies, supernova, whatever you're

measuring, are there ones that you kind of trust more than others? Are there ones that are contributing more to this 4.2 sigma tension, so to speak? From the perspective of the barrenacusic oscillation measurements, I would say we trust all the tracers equally. We have done many, many tests. We've filtered the data. We've taken subsamples of the data. And in none of the individual tracers, do we see any discrepancies or any fluctuations that would be beyond statistical? So on the BEO side, I actually would say all tracers are created equal, all the way from retch of zero to say, like I said, 3.5. The effective redshift of this Lyman Alpha Forest is more like 2.4, but we use those data all the way to the highest redshift possible.

So in terms of constraining power, the sweet spot is more in that mid-redshift range, something like 0.4 to 1.1, 1.2-ish. That's where we have the enough tracers to basically saturate the cosmological information, and we have enough cosmological volume to make really precise measurements. But in terms of the robustness of the data, I would say it's fairly equal across the board. And that's actually one of the things I'm most proud of in this experiment is that what we've attributed to systematic errors in the measurement are way, way, way subdominant. We've looked at it across tracers. And I maybe think that our estimates are conservative. I think they're better than what we've said they are.

So when we hear about 4.2 sigma tension, break that down for the listeners and viewers out there. What does that mean, sigma and so forth, don't be afraid to get technical. The 4.2 Sigma Own it all. Pay off your home, travel for life, drive a Ferrari. In celebration of the world premiere of the Monopoly Big Board Buckslot machine by Aristocrat Gaming, Yamava Resort and Casino at San Manuel

is giving one person a $1.6 million dream package. The biggest prize in Yamava's history. Club Serrano members can earn daily instant prizes and secure a spot in the finale May 29th. Don't pass go and own it all. Only at Yamava, celebrating its 40th anniversary. You win? Details at Yamava.com must be 21-20.

Please gamble responsibly. Monopoly is a trademark of Hasbro. is not a sponsor of this promotion. Tension is a comparison between the model that best describes the universe in a Lambda CDM context and the model that best describes the universe in a more flexible context, one that specifically allows dark energy to have an equitia state that varies with time. And it's really just a comparison of the quality of the kai square between the two fits. So for like particle physics experiments and also now for cosmology experience, we consider

the gold standard for absolute new discovery to be at the 5 Sigma level. If I look at something to the 3 Sigma level, we've had many claims at that level in cosmology, we have tensions at the 3 Sigma level, and they've come and gone. You could kind of take a coin flip if it's real. So 4.2 Sigma is definitely in the regime where you're now starting to really see there's legitimate tension. It's getting close to that 5 Sigma discovery threshold. But I want to air a little caution to the audience. 4.2 Sigma is one particular data combination. It's using one particular supernova data sample. If I swap in a different sum of a sample and extract exactly the same physics, that 4.2 sigma changes. It now goes down to something like 3 sigma or 3.8 sigma, depending on what

supernova sample I'm using. So while I say the BIO measurements are totally robust, we only have one project out there that's produced things at this level precision, where some of the other probes, there's actually multiple samples and multiple analysis techniques at comparable precision, and they disagree with each other a level that changes the tension we see with the Lambda CDM model, at a regime that goes from fairly mundane to being like borderline new physics. So you may not know, or maybe Dan might have told you, but this office used to belong to Jeffrey Burbage, who was a proponent of the quasi-steady-state universe. He did not believe in the steady-state universe. His wife, Margaret Burbage, was a renowned astronomer, and this is one of her plates of a galaxy,

and I have some spectra that she's made as well. This is actually a photographic emulsion that was used to, you can see there's a smudge there. That's not as much as much as much as an actual galaxy. So when we look at these objects and we look at the spectra, how much has changed and how much can you say to people like Jeff or the ghost of Jeffrey Burbage and Margaret Burbage, unfortunately, no longer with us, but how much can you say about the veracity, the reality of dark energy? Are any of these things casting down onto dark energy, as many people, not just Jeff and Margaret and their colleagues, including Fred Hoyle, prominently, and past guest, Jaya Narlaiqar, and many people don't know that he's still alive and he was an acolyte of Fred Hoyle as his advisor?

How much can we say that the Big Bang occurred for real, for sure, quote unquote, and how real can we ascribe the existence of dark energy to be? I find very hard to believe we'd have a universe where there's no big bang and no dark energy. Even the models we're finding where there's some tension, we still have a very significant dark energy component. It still has a very high fraction of the energy content of today's universe. It's still driving a lot of the expansion industry over the last several billion years. We're actually seeing fairly small perturbations.

Like, for example, the distance-reg relationship changes by something like at the 2% level. from what we would see in a Lambda CDM-only universe. That's a level of tension we're seeing. So I am pretty comfortable with dark energy as being a primary energy component. I'm definitely comfortable with the Big Bang. I might have asked the question to my colleague here, who's more on the theory side for alternatives. But observationally, I find it hard to believe

that either two things would not exist. I'm not to Steelman the alternatives, Dan, to the Big Bang, or to Dark Energy. There are people, Sarkar, other people at, say, eminent cosmologist who claim it doesn't exist, didn't happen. I had Rajesh Gupta on last year, claims the universe 26 billion years old. By the way, Burbage and Hoyle had a very, what some people call, preliminary model of dark energy called the creation field, sea field that created matter,

didn't create dark energy. But I had to create matter, which has energy. So, Dan, what do you say to steal now? Not to take it down, but is there any credulity to the claim that perhaps the Big Bang didn't happen? Or perhaps it did happen, but the universe is very different than a universe of Lambda CDM. I think it's always hard to interrogate these questions without a clear model. Because, I mean, the data clearly prefers, like, when you allow it to be flexible, as sort of

Kyle described, you say, okay, let me just, like, be as open-minded as I can and ask what is the data prefer. It generally zooms in on the Lambda CDM model with kind of, again, a lot of dark energy. And we can look at many different probes. So I think, again, as Kyle referred to, this idea that, like, you would trust one measurement alone as being your definitive standard is probably not right. But when you see, okay, like I look at how galaxies form and that also needs to have the dark energy. Like that was, I think, a huge part of the history of the subject was that in the 80s, there were sort of all of these issues with understanding formation of structure. And it kind of not agreeing with this universe that people thought were going to be all matter.

And there's no dark energy. And they kind of always knew that it wasn't working until, you know, we made this measurement of the acceleration of the universe. And then it was like, oh, obviously we should have been doing this all along. Everything fit better when we put that in. And there were people already arguing for the need for dark energy from a completely separate point of view, which is explaining why there are clusters of galaxies and galaxies with the properties that we tend to observe.

And so in general, like, when you give up, it's not that there's just one data set that came in and like everyone else was totally fine with the model of the universe and it was needed. It was really that once we realized that the data preferred that from the supernova, we accepted that actually all the data had wanted that all along, and we had kind of ignored that. And one of the people who I think Nick Kaiser was a very eminent cosmologist during that time, and he used to always say one of his big, you know, I think he took it very hard that, you know, he worked on this study of galaxies, and he would have said they should have been the ones to discover it.

That it was like the fact that they had this thing staring them in the face and they let the supernova people make the discovery. It was like, well, how are we ever going to discover anything if we had this giant signal that we were staring at? We couldn't see it. So I think that's for me, that's sort of the evidence. Now, if there's some model that no one's thought of before that makes everything work just as well, but explain some of these other tensions, like I think we'll all, you know, shift our point of view. But in the absence of a model that really clearly works as well, it's kind of hard to, to, to see how you would take a different point of view. Is it true, Kyle, that as our colleagues, my friends and colleagues, Suzanne Staggs,

Mark Demplin, Lyman Page, I've demonstrated David Spurgel very clearly that the Lambda is unavoidable, or some version of dark energy is unavoidable using the CMB alone? I think that's true, right, Dan? Is it also true from BAA alone? You can derive the imperative of dark energy's existence. Yeah, that is definitely true for BAL alone as well. Our models, if we were to throw out any version of dark energy, would basically be impossible to describe the measurements we see over the

ratio range we make the measurements. We see a preference for something like 70% of the current energy contents being dark energy, and that's hard to get around. Now you parameterized things using this formalism of W0, W-0, W-A plane. Can you talk about equation of state, the limitations of a single parameter encapsulating all this physics, and then what are W-0, W-W-A? What is there a utility in a metric sense. I look at how the energy density evolves with time, and the first order, at least from the Freeman equation, from the expansion industry, that's what that equation say it's going to give me. For a specific model of dark energy, that that comes from Einstein for the constant constant constant,

that energy density would be constant over all times. So that fits the data extremely well. And of course, if you imagine the universe expanding and you have one component that's constant, where everything else gets diluted due to expansion, of course, at later times that energy component becomes dominant. And that is what the data look like. That's what we see. It's like at the 70% level. Now, you can imagine getting more flexibility of that, letting it evolve with time. So the energy density is no longer constant, but it can vary. Specifically, the equation state relates to the energy density to the pressure. But again, thinking about it in terms of the variation with time, what we would be seeing is a component that looks, you know, the data when we combine with

Subinova and C&B looks like this no longer constant with all time at the three to four sigma level, but instead it's kind of decaying to look more matter-like as you get into the local universe, like right-shift zero. So the equation state is evolving for something at higher redshift that looks Einstein-like or cosmological constant like to something today that looks more matter-like, which means it will be, the energy density will be diluted due to expansion as we go to future times. A natural question that people ask when they think about this, if it's evolving, what's causing what type of material would this equation of state represent? I think I have to go to my local theorist.

I think the challenge is just within this parameterization, the regime that has most of the probability of explaining the data is one where it looks like a free lunch. So that, you know, we can just borrow energy from the universe. And yeah, we can just get as much negative energy as we want. And so theorists don't like that kind of regime of parameter because it's like, yeah, I mean, it would be nice. It could power the entire Earth off just borrowing from the vacuum forever and never have to pay it back. Some of my audience will propose it.

So if the machine to capture that energy has not been patented yet, we should stop the podcast and get on it. But yeah, so that's in the specific W-N-W-A parameterization, the sort of the part that fits the data well requires this kind of what's called null energy condition, violating dark energy, but really it's just like it kind of violates this theorist sense that we shouldn't be able to borrow energy indefinitely, that there's a lower amount of energy that you can't go below. And so energy, when you conserve energy in the universe, it means that you can't just keep going more and more negative. And so that's the reason that your summer starts now with Memorial Day deals at the Home Depot. It's time to fire up summer cookouts with the next grill for burner gas grill. on special buy for only $199.

And entertain all season with the Hampton Bay West Grove seven-piece outdoor dining set for only $499. This Memorial Day get low prices guaranteed at the Home Depot. While supplies, price invalid May 14th or May 27th, U.S. only exclusions apply. See Home Depot.com slash price match for details. Curists don't like it. But again, I think the challenge is always that this is a parameterization.

And then the question is, are there paramatrizations that are more theoretically acceptable that still pick up that expansion history. And I know there are lots of theorists who are working hard trying to avoid those things. And again, within any model, there's still some preference for some of these evolving things. But I do think, I think even the DESE paper suggests that the, what they call phantom dark energy is the one that looks most like what you need to fit the data. Thank you, Mr. Robert. What is that? So phantom is this just, it's just another word for this, you know, that evolving in such a way that it looks like you are violating these sort of, there's no free launch kind of conditions for dark energy.

Is it mandatory that it would have to decrease over time in a phantom or the phantoms that would increase dark energies? You know, the equation of state parameter become more and more negative rather than more and more slightly positive. The key is just the phantom dark energy typically just refers to, is there a place where the equation of state allows more negative energy than you would like? There could certainly be evolving dark energy that evolves, that maybe, again, helps with the data somewhat but doesn't violate these conditions. It is true also that there are ways to

violate, they'll allow certain very, very special kinds of negative energy sources that can't be produced. So they're like, in string theory, there are some sort of objects that are these sort of, they ultimately behave kind of like boundaries that are known to have these. They look like negative energy, but because you can't create boundaries out of nothing, they're either there or not. It doesn't, you can't steal energy from them. And so they're not, those kinds of objects are not pathological. And so there could still be ways to make it all work, but just, you know, the theorist gut instinct is that anything that kind of looks like this phantom dark energy that's needed is kind of on the wrong side of theorist prejudices, let's say.

Kyle, in the beginning you mentioned this kind of synchronization of cosmic clocks or these rulers were starting to grow with the inception or the ending of inflation. How critical is inflation to the BAO story? If inflation did, if God tells you tomorrow, sorry, Kyle. there's no inflation. A bouncing universe, there's some other model. Is that critical? Does that change any of the conclusions or theoretical prejudices? Without inflation, you would have to still have some mechanism to produce this standard ruler

or this preferred feature we see in the clustering of matter, which is there, no matter how you slice it, it is there. So there has to be some era of the universe where you have baryon photon coupled plasma. I would be fine to erase inflation if you have a model that produces that preferred feature. I don't know what that model is. I really like the very specific start date to the propagation of the sound waves when inflation ends. That makes a lot of sense to me. If someone can give me an alternate explanation and explain, you know, how we see this feature that seems to be standard at all red shifts, I'm fine, but I don't see that model that's very clearly.

There's more than one discrepancy in the data from previous measurements, so to speak. When we look at the, you know, combined set of tensions, you know, again, think of a your favorite crank that sends you an email, you know, once a week, that, you know, Professor Dawson or Professor Green, you know, I got this new idea. And if you, I'm not good with math, but if you help me with the math, you know, I'll share, you know, I'll share some of the money. I've gotten all sorts of things. You get a Nobel Prize or a fraction of it, or I'll get the Nobel Prize and you'll get the dollars. What do you say to these people that, look, come on. This is crazy. You've got all these tensions. You don't agree on the Hubble constant, you know, five sigma discrepancy.

You need this phantom material called dark matter. We have no idea what it is. dark energy, we have no idea what it is. You do the calculation from first principles and the most embarrassing result in all of science, 120 orders of magnitude. What do you guys really do on over here? This is nuts. How many tensions can we say before cosmology? Again, it's not a serious subject, and we should all be, you know, doing something useful like screwing in iPhone parts. How do you react to that, Dan? How do you react to that first, Kyle? I would argue from an observational perspective, these tensions might be arising because they're real. They're real cracks in the model, which is super exciting.

But it also is a statement about how precisely we can make these measurements. We'd only be seeing things at this level if we can make sub-percent precision measurements, which we're able to do, right? Our aggregate precision in BAO is like 0.3% of measuring distances. And it's hard to measure the diameter or the area of this room at that precision. Measure your age to a couple days, precision, right? Yeah, I mean, if you, well, or you can measure the time since conception. Right? And that is probably impossible to measure at that level.

Yeah, our own Big Bang is not possible at that level of precision. I'm observationally super proud of the fact that we're able to make these measurements as precision, and I would go from there. So the only reason the question would be is the model wrong, or are we starting to see cracks in our ability to make these measurements? Are we starting to see the limits, whether in the CMB or the BAO technique and large-scale structure or in the supernova, are we seeing limits of our current observational technique? Dan, how do you rank these tensions?

I would say my attitude for most of these things is that, again, going back to this idea that a single experiment isn't how we tend to view science, that you want the self-consistency of many experiments. And so, but typically going between experiments required some model. If I look at a dataset just in isolation, then I can make lots of things work, but then take that what you learn from that experiment, ask, what does it teach us elsewhere? So I think in the context of like the Hubble tension, when the people first started talking about it, I was very interested because I thought, oh, well, we actually know that the tension between the BAO measurement of H-0 and the expansion rate and the supernova one

was calibrated off of the cosmic microwave background. So maybe there could be like new physics that explains why there's actually nothing wrong because you calibrated two experiments off of each other that had the same sensitivity to some new. effect that maybe you could make everything work and it was just a discovery of exciting new physics. And the problem is that people have tried that now for a decade and every idea you have for how that would work is immediately ruled out by some other experiment. And so every attempt to make that work, you're like, it sounds great, but then you're like, oh, but when I change

something by that much, it starts just poking out everywhere. And there's no good idea so far. So my attitude towards the Hubble tension is given how hard it is to find an explanation that would make it work, it seems like at some point the most plausible one is some effect in the supernova data or the distance ladder that leads to that measurement that we just haven't figured out what's wrong with it yet, but that still seems more plausible given that there's sort of no plausible reason that you could make them all self-consistent. But when it comes to like, let's say, CME and DESE is actually different, right? I think the tension that we see, while I'm not a big fan of the dark energy interpretation, it is clear that

the tension could be resolved, potentially makes everything fit between those two with either something that's systematic between the two, but also could just be new physics. And new physics is not probed by other experiments. So that's the kind of where, if I, if, you know, as he comes along, this is why it was exciting for neutrino mass, because we couldn't tell how the sort of structure formed in the universe accurately enough to decide until Desi came. And so if there's something interfering with that process or involved in that process, it wouldn't have been ruled out by any other experiments. And so that's why, you know, for now, it seems like they're plausible ideas,

both potentially just to do with some big systematic we didn't understand and some experiment, but also if everyone agrees, there's still totally plausible physics that could explain it, that that's outside of what's going on with the Hubble tension, the supernova. So I think those two I see as being like, it's completely possible. There's something really interesting going on there. But I kind of always want to stick the supernova off the side because I just don't see how to make them fit into a picture that makes sense. All right, Kyle, we've discussed data, but let's talk about the big picture implications.

What could this possibly mean for not just, you know, our branch of cosmology, but possibly for implications for beyond the theoretical realms that we study, maybe even in philosophy? We actually didn't talk about that. Do you want to say anything about that? What are the philosophical implications of a change? changing dark energy or cosmological constant, not being constant. First of all, I actually kind of share Dan's opinion that we should interpret this only within the dark energy context.

What we do to parameterize this, we assign all the flexibility in the expansion history to dark energy. That's where we give it the freedom. But what I really see this is, both more philosophically, is as kind of a null test against the LAM-M-M-M model, the standard model we've had in place for 25 years. And it's starting to see evidence that there's something wrong. with that model. And there's some transition, some perturbation, some extra component that we need to add to the standard model to create the expansion industry in our theory that we see in the data.

So we attribute it all to dark energy and time evolution there, but really it means that we need some component to give you time evolution in the overall energy context, more like the effective equated state for all of the energy contents from the time of the CNB up until today. So philosophically, I actually see that as more exciting as there really does seem to be evident for a need to adapt the overall model. Yeah, I told my students today when I advertised your talk, I told them we might be living through the same type of revolution that we witnessed in the late 1990s when I was in grad school

and you guys were in preschool or whatever. But I remember that being, you know, 1998, the discovery of the year in Science magazine, the picture Einstein blowing bubbles. That was a paradigm shift, right? To what degree are we confident maybe our students are living in the paradigm shift era right now? I mean, we're definitely living in big change

from, let's say, the late 90s through probably 2015, so about 20 years, where it seemed like lots of different pieces of data all were totally consistent with each other. And the students could say, like, go through their life going, the C&B is 100% right. I have, no, I'm not worried about it at all. It clearly not. Yes. So, but yeah, but I still that it deserves. But one way or the other, the cosmology student today has to live with a level of uncertainty about what, you know, what data sets are trustworthy or what they, when they're thinking about how they view the universe, it's not one one coherent picture. And when I had hoped the Huffle tension, for example, would be resolved by now and that DESE would be totally consistent, that was sort of, I was perfectly

happy to have a consistent universe. But yeah, this generation of students are having to learn how to deal with this and we'll presumably have to deal with it for at least another decade, if not longer. So we're living through this era of uncertainty. And that's, you know, that's not a, that's maybe not a paradigm shift on the theory side, but it's definitely a paradigm shift in how theory interacts with observations, for example. Let's switch gears dramatically. Let's go from measuring photons to measuring particles with mass, like these chunks of matter that you'll get if you go to Brian Keene.com slash list and sign up for my Monday magic mailing list. You've got one of these cards in the mail if you are one of the lucky winners, but you're guaranteed to win if you have

a .edu email address like these two gentlemen do. And so Dan, uh, so Kyle, Dan, you already have one, Kyle, you're going to get one. It's a real honest to goodness meteorite chunk of matter, which has, honest of goodness, yeah, nice. You can hear it. I'm talking, of course, about the implications for neutrino mass. And some of the mass limits are getting uncomfortably or comfortably for some, a gentleman to my left. Close to the floor set by laboratory measurements and other measurements. Let's talk about neutrino mass measurements. First of all, Kyle, how do measure neutrino mass, so the sum of neutrino masses, what do you measure about neutrinos? And then we'll pivot to Dan for interpretation, possibly, about their existence of the existence of negative intrinom mass.

And we're going to clear the table and have a little armist. No, we're not going to have arboristly. But Kyle, how do we measure? Ambition comes in all shapes and sizes. At First Citizens Bank, we roll with your goals because we're built for what you're building. Fit for your ambition for Citizens Bank. neutrino mass and how does it pop out essentially for free? And what did you guys make of it when you first started to encounter the results? Neutrino mass affects two things in cosmic evolution,

it affects the growth of the structure, and it affects the expansion rate to the geometric component. Desi, with our BIO measurements, is really able to tease out the geometric component because neutrinos are at the time the time of the C&B decouples. They're a low enough mass that they're still relativistic at that time period. And they become non-relativistic at later times. So they basically transition from being like a photon-like or radiation-like component to a matter-like component into the free-meat equation. They have an equation of state that responds accordingly. So by comparing the DESE measurements to the CMB measurements, we can really get accessed in the geometric contributions from that transition. And what we see is that the expansion rate prefers a mass.

If I work within Puyall-LAM-CD model, that's as low as possible. One could say I have a prior where the mass has to be greater than it equal to zero. And I get a likelihood into the Lambda CDM perspective where the geometry says I don't see the non-relative contributions to the expansion history from neutrinos. It's very weak. And the preferred value is basically M equals zero. From the geometric side, where that really is coming from is that the expansion history that we see in the data is not really aligning well with what that Lambda CDM high-nutrino mass

prediction gives me. I can correct that by allowing the neutrino mass to go negative, and that's just giving more freedom to the equation state or the expansion history. And I get a benefit to the data with the negative neutrino mass. The way I see that is more evidence that our standard model is not describing the expansion history properly. I can resolve that, as we've already discussed, by bringing in the time-evolving equation state for dark energy.

But to me, these two are equivalent. The negative neutrino mass, the time-evolving equation state of dark energy, are really both saying the same thing. the geometric aspects of the model are not aligned with the data, and I need some modification to the expansion history. Dan, you're known for your tweet storms. We had you on to almost two years ago now discussing an epic thread of breakthroughs in the past, you know, 40, 50 years to counter some of the, you know, kind of people that tend to diminish and denigrate the contributions of both theoretical and experimental physics. We won't name names or anything like that. But the point

being how much progress has been named, you encapsulated in a tweet thread heard around the world. And in a recent tweet thread, you had some more or less critical words for the interpretations, at least, the evolving dark energy perhaps, and or the evidence for neutrino mass being negative. And you actually use that preference for slight negative neutrino mass as a reason to be cautious. Explain that. Explain why you should be cautious, especially given that I made a video about negative neutrino mass based on your previous work, which is so fascinating to me, and showed you and not the clown mask that your kids gave me the picture of it. Tell me, why should we be a little bit cautious? And then what would be the implications of a negative neutrino mass? Is it real? Is it

systematic? Okay, ready to come down. I think, again, anything that I would ascribe as real physics in the data, I'd like to see that it resolves all of the loose threads, right? That things are kind of poking out all over the place. And if you want to resolve that, then presumably your resolution should make everything kind of come back together in some nice way. And so what does seem to be the case is that if you look at the DESE results, it suggests a difference in the measurement of the amount of matter, right? That it's like the CMB seems to want more matter and DESE wants less matter. And so the dumbest possible interpretation of negative neutrino mass is that it's like, it's negative matter. And so it solves all the problems.

Well, now, so it doesn't solve all the problems, but it gets back to that point. So at that point, you could start from, well, is it just a calibration issue, right? If both measure matter and they kind of don't agree. Now, the problem is that neutrino mass, what makes the neutrino mass measurement interesting is that neutrinos are an example of something that behaves, like, not like matter for the C&B, because they're so light that they look like radiation. And then they start behaving like matter later in the universe after 380,000 years. after the Big Bang.

So seeing matter go from being a smaller number to a bigger number is kind of what we're supposed to expect out of these measurements, especially if it was a huge mass, that would be of smoking gun signal. Instead it looks like it does the opposite. But of course, so now you could take multiple points of view. Can we explain, you could try to explain that by explaining why the matter seems to go backwards.

So one strategy just from the expansion rate is to say, what happens if some of the dark matter decayed? So I could imagine, like, I have a bunch of matter, and then it, like, just starts decaying. And so it looked like there was more matter before and now less matter. And so that's an idea that it doesn't solve all the problems, unfortunately. But it accomplishes that goal. But it kind of shows you that there's multiple lever arms you can imagine changing. If you just think of it as like there's some disagreement between different measurements the same quantity.

What is the physics that could be responsible? Another way that you can get around this. So one thing that neutrino does is that it looks like matter from the expansion rate, but it's too light to kind of fall on itself, so it makes structures not form. And so there's supposed to be less structure in the universe if you have neutrinos around, massive neutrinos, and instead what we see is more structure.

And so, however, if you just add a way to create more structure, so for example, if you just had like a new force that only talked to dark matter made it kind of like more bind together, you could also solve the problem without necessarily evoking dark energy. But I think the thing that I'd like to see more of, and again, skepticism for my own work,

but also for the work of many others, is, you know, does this pull together all the loose ends? So, for example, one of the things that I care a lot about is this phase shift of the BAO, which is another way to look at neutrinos in the early universe.

So neutrinos travel faster than these sound waves, Kyle was talking about, and so what they actually do is they make a signal that looks supersonic as far as the C&B is concerned, and then also as far as the BAO is concerned later. So you're seeing this, like you're supposed to be able to see

the supersonic signal of neutrinos. We got like a sonic boom. Do we get a Turing? You see information traveling fast from the sound? So that's what we see. We see it's kind of like ahead of the sound wave. There's kind of like stuff happening.

And it wouldn't happen if there wasn't something moving fast enough for to make that signal. It's kind of protected by the special property of causality, right? That's like things have to move from here to there to create that signal as you need something to actually move fast in the sound waves. So Desi does this analysis, which I was very excited to see because we had done this on boss data in the past And they do see as evidence for neutrinos, but too much evidence. It's like infinite number of neutrinos, I think is where the fit is. But it's like it's high by almost three sigma in the spaceship. And even when you add dynamical dark energy, it's still too high like two and a half sigma.

And so anything that kind of like brings everything together, you'd like so that would like to, you'd like that to come back to what it's supposed to be. So that can also potentially be resolved. That's most likely again, just another. other symptom of the disagreement between the C&B and BAO. But that's that's sort of what I think a solution would look like, whether it's ultimately explained by a systematics from one of the experiments or all of them, or is ultimately explained by new physics, you would just like all of those things to like one model that's like, yes,

the reason this was high and this was off here is that they all kind of come into place. And I think we just don't have an example of that right now. And Kyle, when we look towards, you know, the famous statement, I think originated maybe by Jim P. Bowles or at least said by. him more than anybody else. You know, interesting if true or something like that or trust but verify. I don't know. How can these results be qualified or confirmed or refuted as we have to admit the possibility

thereof? The results of the BIO measurements. Well, I would say in terms of our precision and our robustness, we really do not see any evidence no matter how we slice the data that there's any systematic errors that are pushing the analysis or the results one way or the other. I would say that it would be nice to see all the data to decide. I guess that question for Kyle was how is Euclid, who's coming fast on your heels, going to inform this question?

And how long until we see? Do you expect before you're competing, Euclid's competing with Dizzy? Yeah, I haven't looked at the data release schedule exactly for how much area they'll have as a function of time. But Euclid is preferentially a higher redshift. And a lot of the information that we really are gleaning and where we're getting the better constraints in Omega Matter, it's because we have a really large lever arm in redshift. So we go down to right shift point four with pretty high precision. Our lowest right shift measurements actually 0.25 or something like that, where Euclid won't have that level of sensitivity.

So I think Eucl would be nice for validating or overlapping our right shift between redshift 1 and 2, say. But that's not where most of our information is coming from. And I think it would basically help constrain the distance measurements in the matter-dominated regime, but be a little tougher to interpret at the level we have at the level we've done. When Adam Reese was in that chair a year or so ago, you know, we talked about the, you know, the saturation of the measurements of supernova, the different surveys that he's been involved with. And, you know, it's kind of, as you mentioned, the point, you know, fraction of a percent level precision. How much can be improved?

And then with DESE as it is, and then tell us a little bit about what is DESE 2. Dezzi, we're pushing to observe as long as possible. What's effective of the same survey strategy up until basically the end of 2028. and we can probe something like two-thirds of the extra-galactic sky. We can get higher number density of tracers, so we're getting pretty close, at least out to reach of one, to the cosmic variance limits. We're not really limited by number density.

And we do not see systematic errors as reducing it. So it's all data volume that gets us the better precision. We're not losing, we're going to go with the data volume. At that point, we start running in a sky. We can't observe further south than some declination. We can't get any many more galaxies because we're so far beyond the shot noise limits. We've already sampled all the modes that are out there for BAO, roughly speaking. It's not efficient to increase the number density significantly.

So we're going to push to that limit as hard as we can. And what about, so DESE 2 is that in proposal phase? Where are we sending? DESE 2 is still in the proposal phase. And the biggest change we're hoping for DESE 2 is to move to a whole new retrof regime. And we've actually been talking about DESE 2 for longer than we've. we've known that dark energy might not be what we thought it was. So we've actually, since snow mass, the last, actually arguing from my play last seven or eight

years that we need to move beyond dark energy, ironically, and get to more early universe physics. And the best way of sampling the early universe physics is to get tracers that are higher redshift, where you have less processing of the matter density field, you have more volume, there's a lot of information content that we're unable to pull out of the data right now. So our goal has always been to get to kind of change the role of cosmology away from dark energy into more inflationary physics and other type of physics. So that's what the goal of DESE 2 is, is a stepping stone towards that new future. Okay, let's hit pause for a second. We might be having our heads spinning around from all these different talks, models being right, models being tested, models potentially being overthrown.

And so if your brain's spinning right now, you're not alone. We have a model, Lambda CDM, that's been the cosmic standard for 25 years since I was a wee graduate student. Now, multiple data sets from different directions, from different instruments, from supernovae to barrier on acoustic oscillation, the CMB, might be telling a story that's a lot different than the one we told ourselves back 25 years ago at the end of the previous century, hard to believe, millennium even. And the real kicker is that Desi may be providing evidence that these tensions might not be getting better. They might be getting worse. We might need that cosmic therapist that I've been advocating for so long. So coming up, we're going to ask Kyle and Dan something really spectacular and something bold. if these results hold up, what does that mean for the future of cosmology and even the future

of the universe? What would happen in the old take us out a trillion years? We still don't know what the evolution would be. I mean, I think that puts back on the table a lot of future. I mean, we've, we've, the equation of state could allow for all kinds of things, including one, like, I think. When you need to build up your team to handle the growing chaos at work, use indeed sponsored jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications, and more. Spend less time searching and more time actually interviewing candidates who check all your boxes.

Listeners of this show will get a $75-sponsored job credit at Indeed.com slash podcast. That's Indeed.com slash podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed sponsored jobs. Back to all the ending scenarios. The universe expands forever or crunches. I think we don't know enough to say if we don't know what the evolution is, a trillion

years from now, it could do all kinds of things. I think we just don't have the lever arm. Once you open up that freedom, it could be basically anything. And so we could all die in a giant crutch. The heat death is off the table. Is that true? If it's, again, assuming this is correct, their results stand up and confirm. I'll leave that to Kyle.

Well, I don't know if I want to get into that. But what I will say, I think what Dan is getting at is this parameterization we have for dark energy. allows a lot of extrapolation into the future, which is what you're asking, and a way that doesn't necessarily line up with any good model. So it's really a pretty phenomenal way

of characterizing the creation state. It's really built to be local, like what's happening about the evolution of dark energy in the dark energy-dominated regime. I find myself a little bit skeptical when I extrapolate that to these earlier times the universe, like this idea of the Phantom Crossing and this earlier times,

that's not where our data, at least from the BEO side, are really probing. That's coming from the contrast against the CMB. I find it equally uncomfortable projecting the future, because if you take this linear extrapolation and go forward, you're going to get to an equation of state that's like infinitely positive, right? If we took it as naively as possible, which is clearly not going to happen. So what the next step really has to be is to get to higher precision measurements, where we can really see the time evolution of equation of equation state,

not in this parameterization, but in a way that allows us to probe specific models and maybe hone in a model that's more physically motivated for describing what we're seeing. Can I add? I think one of the things that that tells you, though, is that these conclusions or even when we, again, used to say with certainty, we know what the future of the universe is and it's the thief death. It told you that was always anchored in our belief in a model and that if we give up our belief in a good model that we would trust beyond the data, right? That this is a good reason this model is the right model, And yeah, the future is just back on the table. Everything's on the table because we can't extrapolate without a good model.

Yeah, I think what I would say along those lines, if we get a five-sign detection and the LAMC-DM is wrong, I would say we have no model. That's actually my interpretation of the results. We do not have a model for cosmological expansion, cosmological history. So soon we're going to talk about the deep implications for philosophy, even people have speculated what these might mean for religion, for theology. We'll get to those.

But I do want to ask Kyle, especially, about the technical details because we had these photographic plates for, you know, almost a century. And yet this project is sponsored in part by the Department of Energy, right? Why is the Department of Energy interested in this? What is the role of big data, machine learning? Could this project have been done 25 years ago? Or is this now? Why now is because it only could be done now? So last way to your question.

First, why is the Dark Energy, why is Department of Energy interested in this? They have a long history of sponsoring particle physics experiments and really precise. decision measurements of our underlying model for all the physics. And I would say this measurement of dark energy, understanding cosmology, really fits into that, that whole umbrella. So that's where that fits in. In terms of why now, this absolutely could not have been done 25 years ago. We did not have the technology to multiplex or spectrogosphi at this level. We did not have the robotic positioners that could center the fibers at this level precision. We didn't have dedicated collaborations. We didn't have the software to build up the simulations and the

mocks. We had hints of it. We had the first discoveries, the first measurements of the barrenica's collation almost 20 years ago on the clustering of matter. So we had the first ability to make 3D measurements of the clustering of matter. But at this level precision, it was, this is now the, roughly speaking, the third generation project developing the technique. And it's only up to now that we can do it at this precision we're talking about. The first generation was the discovery that was the Solend Judo Sky Survey. And it was also, there's other projects that were at the similar scale. The second generation was the slinger skies survey with the boss spectrograph, which we upgraded to have more power than the

original instrument, the original spectrograph. That allowed us to go to higher Redshift, to get spectroscopy out to Redshift almost one with galaxies alone and also at higher multiplex. And then the latest generation, which is DESE, allows us to get up to 5,000 objects with a much larger mirror. We're going more than 10 times faster than we were able to in boss and e-boss. And it's only with that level of technology that we're able to do what we can do. This has been fascinating as usual, and there's a lecturer, or actual colloquium. And there's a saying, at least I learned from my late, great colleague Hans Parr, who said, you know, if you measure something at 3 Sigma, you'll get invited to give a seminar. If you measure something at 4 Sigma, you'll get invited to give a colloquium and we'll pay for your visit.

And 5 Sigma, you get invited to Stockholm, you know, essentially. So, you know, you might get one of these. These are actually from Stockholm. These are, I won't tell you how old. These are, God forbid, someone tries to eat one of these. Rest in peace. Has been fascinating. Kyle Dawson, thank you.

you for coming all the way from from Utah to give the talk. I can't wait to hear it later on today. And Dan, thanks as always for making the journey from, uh, from mayor hall down here to surf. And, uh, we'll be doing many more of these in the future. So stay tuned. We have upcoming guests, including David Wiltshire. So David Wiltshire is a proponent of the so-called timescape cosmology. And he's been, uh, frequently requested guest. Uh, maybe we'll talk about that over lunch, which we owe you starting now. Thank you so much, Kyle. And then. Thanks, Brian. Thanks, Brian for my first podcast. Enjoyed it. Me too.

If you want to understand the theoretical underpinnings of dark energy and why it might not be a cosmological constant, you want to dive into this episode next with theoretical physicist Eric Weinstein's captivating talk at UC San Diego. Join Eric on a discovery of why Einstein may have been wrong. Click here to watch that. There's a new way to sweet green. Meat, wraps, handheld, hearty, and made for life on the moon. With bold, chef-crafted flavors, fresh ingredients, and over 40 grams of protein, They're built to satisfy without slowing you down.

Try wraps today in the app or at order. Dot sweetgreen.com, available at all participating locations.