Historical research guide · 0397 · 2025

DESI DR2 Results II: Measurements of Baryon Acoustic Oscillations and Cosmological Constraints

DESI Collaboration · M. Abdul Karim and 185 co-authors

In one minute

The Dark Energy Spectroscopic Instrument measures distances across the sky by finding a fixed ruler printed on it — the baryon acoustic oscillation, a sound wave frozen into matter when the young universe cooled. Here the DESI collaboration reports that ruler as measured in more than fourteen million galaxies and quasars from three years of observing, the largest such sample assembled and the most precise BAO measurements yet made. What the numbers say is the interesting part. A universe with a constant dark energy — Einstein’s cosmological constant, the standard model of cosmology — does fit DESI’s own data, but the parameters DESI prefers sit in 2.3 sigma tension with those from the cosmic microwave background. Let dark energy change with time instead and the tension eases: the collaboration finds a time-evolving equation of state preferred over a constant one at 3.1 sigma from DESI and the microwave background alone, and at 2.8 to 4.2 sigma once supernova samples are added. Their own verdict is that the standard model is being challenged.

Connection

Chapter 2 treats the vacuum as a real medium with an energy density, and chapter 13 puts dark energy and that vacuum energy under one description. If dark energy is changing with time, then whatever fills the vacuum is not a fixed constant of nature but a quantity with a history — which is exactly the measurement this site watches.

From the source

DESI DR2 delivers baryon acoustic oscillation measurements from more than 14 million galaxies and quasars over redshifts 0.1 to 2.1, covering a cumulative effective volume of over 42 cubic gigaparsecs; the precision on the isotropic distance scale ranges from 1.54 percent for quasars to 0.45 percent for the combined luminous-red-galaxy and emission-line-galaxy sample, and these are the most precise BAO measurements ever made at every redshift covered.Section IX, Conclusions, paragraphs 1 and 3
The combined DR2 BAO data are well fit by a flat cosmological-constant model with a matter density of 0.2975 plus or minus 0.0086 and a scaled Hubble constant times sound horizon of 101.54 plus or minus 0.73 megaparsecs — a 40 percent improvement in precision over DR1 — yet those parameters are in 2.3 sigma tension with the ones derived from the cosmic microwave background, up from 1.9 sigma in DR1, even though DESI agrees with the acoustic angular scale the microwave background measures so precisely.Section IX, Conclusions, paragraphs 5 and 6
Letting the dark energy equation of state evolve, in the form w of a equals w-nought plus w-a times one minus a, relieves that tension and is preferred over the cosmological constant at 3.1 sigma for DESI plus the microwave background alone, and at 2.8, 3.8 and 4.2 sigma when the Pantheon-plus, Union3 and DESY5 supernova samples respectively are added; every combination studied favours the same quadrant, with w-nought greater than minus one and w-a less than zero.Section VII A, Results; Table VI; Figure 11
In all cases the favoured equation of state shows a phase above minus one at low redshift and a crossing below minus one above a redshift of about 0.4, a phantom crossing that violates the null energy condition; the companion paper’s binned reconstruction, which assumes no functional form for the equation of state, finds the same picture, so the behaviour is largely independent of the parametrisation chosen.Section VII A, final paragraphs; Figure 12
The collaboration tests the alternatives and reports that they do not dissolve the result: a coherent 1.5 percent systematic shift in all measured isotropic BAO scales would ease the tension but exceeds the measured systematic error budget by a factor of six to ten; letting spatial curvature vary freely does not move the dark-energy parameters back towards a cosmological constant; neutrino mass has almost no effect on them; and replacing the microwave background entirely with Dark Energy Survey Year 3 weak lensing, using only low-redshift data, still excludes the cosmological constant at high significance.Section VII C, Are alternative explanations possible?; Figure 14
The result is not yet at the 5 sigma threshold conventionally required for new physics — no data combination exceeds 4.2 sigma — and the collaboration names what would settle it: uniformly calibrated supernova samples spanning both the low- and high-redshift regimes, from ZTF and LSST, together with sharper measurements from future DESI analyses and other experiments.Section VII C, paragraph 2; Section IX, final paragraph

The way in

https://arxiv.org/abs/2503.14738

Historical source-access record. This guide links to the original; it does not reproduce the paper text.

arXiv:2503.14738v3 [astro-ph.CO], 9 October 2025, published as Phys. Rev. D 112, 083515 (2025). The arXiv record carries a Creative Commons Attribution 4.0 International licence, so the text below is reproduced from the paper itself. The author list runs to 186 names led by M. Abdul Karim; correspondence is to the DESI spokespersons at Lawrence Berkeley National Laboratory. Reproduced here are the abstract, Section I (Introduction), Section VII (Dark Energy) and Section IX (Conclusions), which are the parts of the paper this library is reading it for. Sections II to VI and VIII, the appendices, the acknowledgements and the reference list are omitted for length; display equations, figures and tables are not reproduced, and the numbered citation markers of the original have been dropped. Inequalities are written in words for rendering safety. The complete text, the figures and the data are at the source.

How to cite it

DESI Collaboration, M. Abdul Karim and 185 co-authors (2025) DESI DR2 Results II: Measurements of Baryon Acoustic Oscillations and Cosmological Constraints doi:10.1103/tr6y-kpc6

Where it sits in the curriculum

The vacuum at cosmic scale