Of Particular Significance

DESI Shakes Up the Universe

POSTED BY Matt Strassler

POSTED BY Matt Strassler

ON 04/05/2024

It’s always fun and interesting when a measurement of an important quantity shows a hint of something unexpected. If yesterday’s results from DESI (the Dark Energy Spectroscopic Instrument) were to hold up to scrutiny, it would be very big news. We may well find out within a year or two, when DESI’s data set triples in size.

The phrase “Dark Energy” is shorthand for “the energy-density and negative pressure of empty space”. This quantity was found to be non-zero back in 1998. But there’s been insufficient data to determine whether it has been constant over the history of the universe. Yesterday’s results from DESI suggest it might not be constant; perhaps the amount of dark energy has varied over time.

If so, this would invalidate scientists’ simplest viable model for the universe, the benchmark known as the “Standard Model of Cosmology” (not to be confused with the “Standard Model of Particle Physics.”, which is something else altogether.) In cosmology’s standard model, nicknamed ΛCDM, there is a constant amount of dark energy [Λ], along with a certain amount of relatively slow-moving (i.e. “cold”) dark matter [CDM] (meaning some kind of stuff that gravitates, but doesn’t shine in any way). All this is accompanied by a little bit of ordinary stuff, out of which familiar objects like planets and bloggers are made.

While ΛCDM agrees with most existing data, it’s crude, and may well be too simple. Perhaps it requires a small tweak. Or perhaps it requires a larger adjustment. We have already had a puzzle for several years, called the “Hubble tension”, concerning the so-called Hubble constant, which is a measure of how quickly the universe has been expanding over time. Measurements of the Hubble constant can be made by studying the nearby universe today; others can be made using views of the universe’s more distant past; and the two classes of measurements disagree by a few percent. This disagreement suggests that maybe there’s an important detail missing from the standard picture of the cosmos’s history.

Now, perhaps DESI has seen a sign of something else in ΛCDM breaking down… specifically the idea of a constant Λ. (At the moment, I know of no obvious relation between DESI’s results and the Hubble tension.) But it’s too early to say for sure; in fact, even if DESI’s results hold up over time, it might be that there are multiple interpretations of their results.

An aside, in answer to a common question: Is the whole concept of the Big Bang at risk? I doubt that very much. The discrepancies are only a few percent. They seem enough to potentially challenge important details, but nowhere near enough to undermine the whole story.

There are a number of murky elements here; in particular, DESI’s data is still limited enough that one’s interpretations of their results depends on one’s assumptions. (There’s also a story here involving the masses of neutrinos, which also depends upon one’s assumptions.) I don’t understand all the issues yet, but I’ll try to wrap my head around them soon, and report back to you. It happens that I’m in the middle of some travel for talks about my book (in CA, WA and OR; you can check my event page) so it may take me a little while, I’m afraid. In the meantime, you might want to read about the Baryon Acoustic Oscillations [BAO] that lie at the heart of DESI’s measurements.

[Postscript: I guess Nature liked my title and decided to shake up New Jersey today… as felt today even in Massachusetts.]

Share via:


15 Responses

  1. I certainly cannot equal the deep understanding exhibited by the majority of your commenters. Nevertheless I have had some training and do have a fairly deep historically-based understanding of how such dilemmas get resolved. When, as was the case with the invariant [cosmic speed limit] and the quantum behavior of very small things, there was sufficient anomalous data, even though it took a few years, matters did get resolved.

    1. Over time, things do get resolved; that’s the glory of science. But the immediate question here is whether DESI’s seemingly anomalous data will stand tall over time, creating a dilemma that will require resolution, or whether the data will turn out to have been just a statistical fluke, leaving no dilemma at all. To answer that question, we need more time and more data. A year or two.

  2. Dr.Strassler:
    Can you elaborate a little as to why dark energy is referred to as “negative pressure”

    1. Well, first of all, “dark energy” is not referred to as “negative pressure”; aside from the fact that such an identification isn’t quite correct, it also has the logic backwards.

      There is a combination of positive energy-density and negative pressure which can be present in empty space without violating Galileo-Einstein relativity. (i.e. it does not define a preferred rest frame, and it does not pick out a particular spatial direction.) This is what was observed to be non-zero in 1998.

      This combination is referred to, in shorthand, as “dark energy”. Not the other way round.

      So the question is: why is positive energy-density and negative pressure referred to as “dark energy”, even though that’s not what it is? Again, shorthand. No one wants to say “that relativity-preserving combination of energy-density and negative pressure” every time they want to refer to it. So “dark energy” is the shorthand.

      Many things that are used as shorthand in physics and astronomy end up misleading non-experts, because it makes them sound different from what they actually are.
      “Dark matter” is shorthand too. It may not be matter, at least not by most people’s definitions of that ambiguous word.

      For more discussion on this website of negative pressure and its effects, read Thomas Sauer’s questions and my answers to this post: https://profmattstrassler.com/2023/09/23/mass-weight-and-fields/

      1. Dr.Strassler:
        I guess what I’m conflating is the following:
        If I had a box of an ideal gas, at some temperature, there would be a “positive pressure” trying to push the box apart. If I was to add energy to my box of gas, let’s say by heating it, the positive pressure would increase, trying to push the box apart even harder. However, I guess from a general relativity standpoint, this “increase” in energy density, would increase the gravitational mass of my box of gas, trying to contract it. However,the increase in positive pressure would overwhelm the increase in gravity, and push the box apart. Isn’t this how stars achieve equilibrium?

        If the energy density of free space is positive, this should increase the gravitational attraction, but it somehow endsup pushing space apart? Is this correct understanding?

        1. The negative pressure beats the positive energy. In ordinary gases, the energy stored in the masses of the atoms far exceeds any pressure, and so the object wants to collapse under gravity. The pressure gravitates, slightly increasing gravity’s pull inward, even though positive pressure pushes against the box’s walls. Here, the negative pressure is larger than the energy stored in the masses of the atoms — it is a highly relativistic system — and so your ordinary intuition about gases does not apply. The energy density creates inward gravity, but the negative pressure creates outward gravity and wins the battle.

          1. Dr.Strassler:
            Thank you. I’m familiar with gravitomagnetism, so I’m guessing it’s something similar to that.

            1. That’s getting too complicated; it’s simpler. If you are already familiar with gravitomagnetism, then you already know far more than you need. Just look at the basic equations for an expanding, uniform universe, known as the Friedman equations, https://en.wikipedia.org/wiki/Friedmann_equations . You’ll see the pressure appears explicitly, and that a large negative pressure causes the claimed effect.

  3. The really interesting find is perhaps that DESI probes galaxies from within the supernova results up to z < 0.15 with Cepheid calibration to halfway the universe history (0.1 < z < 4.2) and yet see the Hubble constant of the most distant probes (H_0 = 67.97 ± 0.38 km s^−1 Mpc^−1). No tension there. And it can lower the neutrino mass sum if dark energy is constant.

    Q: Isn't that the lowest estimate thus far (Sum(m_ν) < 0.072 (0.113) eV at 95 % confidence)?

    But there is a tension when they allow dark energy to vary, since the neutrino mass sum may go below the 3.1ish that it should have. I haven't had time to see if they do a model comparison with having more degrees of freedom. But suggestions of new physics looks iffy based on the DESI results. Instead it looks like it suggests the supernova results are the biggest offenders.

    1. As an afterthought, eROSITA gamma ray quasar study covered much the same distances (and used the DESI data as well) and did a model comparison test using Bayesian formalism: it prefers LCDM.

      The SRG/eROSITA All-Sky Survey: Cosmology Constraints from Cluster Abundances in the Western Galactic Hemisphere

    2. My understanding, Torbjörn, is that they really aren’t sensitive to small redshift (z << 1) even though they do use it. This is because there are just too few galaxies in that relatively small volume, and so there's not much statistical weight from those galaxies in their results. Therefore, even though in principle they do use low-redshift galaxies, their measurement of the Hubble constant is really a higher-redshift measurement... and so they maintain, rather than resolving, the Hubble tension between low- and high-redshift measurements.

      1. Thank you, Matt. My understanding is that they do include the supernova results in the Hubble constant estimate. If so, the supernova results should have their relative statistical weight, and any tension would come only if those results are promoted on their own.

        1. They probably have one estimate without the supernovas and one with; I haven’t had time to look at the papers, between my own book-related travel and the eclipse. It may be next week before I can get back to DESI… but it’s not going anywhere for a while.

Leave a Reply


Buy The Book

A decay of a Higgs boson, as reconstructed by the CMS experiment at the LHC


[UPDATE May 10, 4:00 pm Boston Time, 10:00 pm EU time: AURORAS MAY BE ACTIVE NOW!] I don’t use exclamation marks in blog post titles

POSTED BY Matt Strassler

POSTED BY Matt Strassler

ON 05/10/2024

A couple of days ago, I noted a chance of auroras (a.k.a. northern and southern lights) this week. That chance just went up again, with

POSTED BY Matt Strassler

POSTED BY Matt Strassler

ON 05/09/2024