🔭 astrophysics

Cosmological constraints from the DESI DR1 joint power spectrum and bispectrum analysis

This paper presents the first ShapeFit cosmological inference results using the DESI DR1 bispectrum, demonstrating that combining the bispectrum with the power spectrum significantly tightens constraints on the amplitude of fluctuations while yielding results consistent with standard $\Lambda$ CDM and previous DESI analyses.

Original authors: S. Novell-Masot, H. Gil-Marín, L. Verde, J. Aguilar, S. Ahlen, D. Bianchi, D. Brooks, F. J. Castander, T. Claybaugh, S. Cole, A. de la Macorra, J. Della Costa, S. Ferraro, A. Font-Ribera, J. E. Forero

Published 2026-03-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: S. Novell-Masot, H. Gil-Marín, L. Verde, J. Aguilar, S. Ahlen, D. Bianchi, D. Brooks, F. J. Castander, T. Claybaugh, S. Cole, A. de la Macorra, J. Della Costa, S. Ferraro, A. Font-Ribera, J. E. Forero-Romero, E. Gaztañaga, S. Gontcho A Gontcho, A. X. Gonzalez-Morales, G. Gutierrez, J. Guy, C. Hahn, H. K. Herrera-Alcantar, K. Honscheid, C. Howlett, M. Ishak, J. Jimenez, R. Joyce, R. Kehoe, D. Kirkby, A. Kremin, C. Lamman, L. Le Guillou, M. Manera, A. Meisner, R. Miquel, S. Nadathur, G. Niz, W. J. Percival, I. Pérez-Ràfols, G. Rossi, L. Samushia, E. Sanchez, E. F. Schlafly, D. Schlegel, M. Schubnell, J. Silber, D. Sprayberry, G. Tarlé, B. A. Weaver, C. Zhao, R. Zhou

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Mapping the Universe's "Fingerprint"

Imagine the universe as a giant, expanding ocean. For decades, astronomers have been trying to understand the currents and the depth of this ocean by looking at the waves. In cosmology, these "waves" are the distribution of galaxies.

This paper is about a new, super-precise map created by the Dark Energy Spectroscopic Instrument (DESI). DESI is a massive telescope project that has taken spectra (chemical fingerprints) of millions of galaxies and quasars. The goal? To figure out how the universe is expanding, what it's made of, and whether our current understanding of physics (called the $\Lambda$ CDM model) is correct or if there's some "new physics" hiding in the shadows.

The Old Way vs. The New Way

The Old Way (The "Ruler" Approach):
Traditionally, astronomers looked at the galaxy map like a carpenter looking at a ruler. They looked for a specific, repeating pattern in the distribution of galaxies called Baryon Acoustic Oscillations (BAO). Think of this as a "standard ruler" left over from the Big Bang. By measuring how long this ruler looks to us now, they could calculate distances. This is like measuring how far away a lighthouse is by seeing how big its light beam looks.

The New Way (The "Shape" Approach):
This paper uses a method called ShapeFit. Instead of just looking at the ruler, the authors look at the entire shape of the galaxy distribution.

The Analogy: Imagine you are trying to identify a person in a crowd.
- The Old Way just measures their height (the ruler).
- The New Way (ShapeFit) looks at their height, their build, the slope of their shoulders, and the curve of their back. It captures much more detail.

The Secret Weapon: The "Bispectrum"

The biggest innovation in this paper is the use of the Bispectrum.

The Power Spectrum (2-point): This measures how galaxies cluster in pairs. It's like asking, "How often do I see two people standing next to each other?"
The Bispectrum (3-point): This measures how galaxies cluster in triplets. It's like asking, "How often do I see three people standing in a triangle?"

Why does this matter?
Imagine trying to guess the rules of a game just by looking at pairs of players. You might get confused. But if you look at groups of three, the patterns become much clearer.
In this study, adding the "triplet" data (the bispectrum) helped the scientists untangle two things that were previously stuck together:

How fast the universe is growing (structure growth).
How "clumpy" the universe is (amplitude of fluctuations).

By looking at triplets, they could separate these two effects, which tightened their measurements significantly.

What Did They Find?

The team ran their "ShapeFit" analysis on the DESI data and found some reassuring results:

The Standard Model Still Holds: The universe seems to behave exactly as our standard model ( $\Lambda$ CDM) predicts. There is no smoking gun for "new physics" yet. The dark energy (the force pushing the universe apart) still looks like a constant cosmological constant ( $\Lambda$ ).
Tighter Constraints: Because they used the "triplet" data, their measurements of the universe's "clumpiness" became 20% more precise. It's like going from a blurry photo to a high-definition one.
The Hubble Constant ( $H_0$ ): They measured the current expansion rate of the universe. Their result ($68.92$ km/s/Mpc) sits comfortably between the "early universe" measurements (from the Cosmic Microwave Background) and the "late universe" measurements (from nearby supernovae). It doesn't solve the famous "Hubble Tension" (the disagreement between these two methods), but it confirms that the DESI data is consistent with the standard model.
Neutrinos: They put an upper limit on the total mass of neutrinos (ghostly particles that pass through everything). They found that if neutrinos have mass, it must be very small, consistent with the "normal" hierarchy of particles.

The "Sound Horizon" Twist

One of the coolest parts of the paper is a test called "Sound-Horizon Marginalization."

The Problem: Usually, to measure distances in the universe, we rely on a "standard ruler" (the sound horizon) that was set in the very early universe. But what if our understanding of that early universe is slightly wrong?
The Test: The authors said, "Let's pretend we don't know the size of that ruler. Let's treat it as a free variable."
The Result: Even without assuming the size of the early-universe ruler, they could still measure the expansion history of the universe today. The results were slightly less precise (the error bars got bigger), but they were still consistent. This proves their method is robust and doesn't rely too heavily on assumptions about the Big Bang.

The Bottom Line

This paper is a "stress test" for our understanding of the universe.

The Verdict: The universe is behaving exactly as we expect it to.
The Win: The new method (ShapeFit + Bispectrum) works beautifully. It's a powerful, flexible tool that allows scientists to get high-precision results without getting bogged down in overly complex modeling.
The Future: As DESI collects more data (it's only at the beginning of its mission), these measurements will get even sharper. If there is "new physics" hiding in the universe, this is the kind of tool that will eventually find it.

In short: The authors took a massive dataset, used a clever new way of looking at the "shape" of the galaxy distribution, and confirmed that our current map of the universe is still accurate, while giving us a much sharper lens to look for cracks in the map in the future.

1. Problem Statement

The Dark Energy Spectroscopic Instrument (DESI) Data Release 1 (DR1) provides an unprecedented volume of galaxy clustering data, mapping over 60 million objects. While standard analyses of this data typically rely on two-point statistics (the Power Spectrum, $P(k)$ ) to measure Baryon Acoustic Oscillations (BAO) and Redshift-Space Distortions (RSD), these methods often marginalize over the "broadband" shape of the power spectrum, discarding significant cosmological information.

Higher-order statistics, specifically the bispectrum (three-point correlation), contain additional information that can break parameter degeneracies (e.g., between the growth rate $f$ and amplitude $\sigma_8$ ) and tighten cosmological constraints. However, applying full-shape modeling to the bispectrum is computationally expensive and sensitive to nuisance parameters. The challenge is to extract the full information content of the joint power spectrum and bispectrum in a robust, model-independent manner without the computational overhead of full direct fitting for every cosmological model.

2. Methodology

The authors employ the ShapeFit framework, a methodology designed as a middle ground between standard compressed BAO+RSD analyses and full "direct fit" modeling.

Data Vector: The analysis utilizes DESI DR1 data for Luminous Red Galaxies (LRG) and Quasars (QSO) across redshifts $0.4 < z < 2.5$ $0.4 < z < 2.5$ .
- LRG & QSO: Joint analysis of Power Spectrum multipoles (monopole, quadrupole, hexadecapole for LRG; monopole/quadrupole for QSO) and the Bispectrum monopole.
- BGS & ELG: Reconstructed BAO constraints (from previous DESI work) are used for Bright Galaxy Survey (BGS) and Emission Line Galaxy (ELG) samples, as bispectrum analysis was not performed for these due to systematics or lower effective volume.
- Lyman- $\alpha$ : BAO constraints from the Lyman- $\alpha$ forest are included.
ShapeFit Compression: Instead of fitting cosmological parameters directly to the raw data, the data vector is first compressed into a set of intermediate parameters:
- $\alpha_{iso}, \alpha_{AP}$ : Isotropic and anisotropic BAO scaling factors.
- $f(z), \sigma_{s8}(z)$ : Growth rate and smoothed matter fluctuation amplitude.
- $m(z) + n(z)$ : A combined shape parameter modeling the slope of the power spectrum near the matter-radiation equality scale and variations in the primordial spectral index.
Inference: The compressed parameters and their covariance matrix (calibrated on blinded data using Renormalized Perturbation Theory for $P(k)$ and GEO-FPT for the bispectrum) are used in a Markov Chain Monte Carlo (MCMC) analysis to infer cosmological parameters under various models.
Models Tested:
1. Baseline $\Lambda$ CDM: Standard flat cosmology.
2. Sound-Horizon Free: Marginalizing over the sound horizon scale ( $r_d$ ) to test late-time physics without early-universe assumptions.
3. Evolving Dark Energy: Using the CPL parametrization ( $w_0, w_a$ ).
4. Neutrino Mass: Constraining the sum of neutrino masses ( $\sum m_\nu$ ).

3. Key Contributions

First Joint ShapeFit-Bispectrum Analysis: This is the first work to derive cosmological constraints from DESI DR1 using the ShapeFit framework applied to the joint power spectrum and bispectrum.
Breaking Degeneracies: The study demonstrates that adding the bispectrum monopole breaks the degeneracy between the growth rate $f$ and the amplitude $\sigma_8$ , which is a known limitation of two-point statistics alone.
Robustness Validation: The results validate the ShapeFit compression approach, showing that it recovers constraints consistent with official DESI full-shape analyses but with a more modular and computationally efficient pipeline.
Sound-Horizon Independence: The authors provide "sound-horizon-free" constraints on $H_0$ and $\Omega_m$ , decoupling low-redshift clustering from early-universe physics assumptions.

4. Key Results

Baseline $\Lambda$ CDM

Consistency: The results are broadly consistent with previous DESI DR1 analyses and Planck 2018 CMB data.
Parameters:
- Matter density: $\Omega_m = 0.310 \pm 0.012$
- Hubble constant: $H_0 = 68.92 \pm 0.97$ km s $^{-1}$ Mpc $^{-1}$
Impact of Bispectrum: Including the bispectrum reduces the error bars on the amplitude of fluctuations, $\ln(A_s \times 10^{10})$ , by approximately 20% compared to using the power spectrum alone ( $3.125 \pm 0.079$ vs $3.18 \pm 0.10$ ). This improvement is statistically equivalent to increasing the survey volume by a factor of $\sim 1.8$ .

Sound-Horizon Marginalization

When treating the sound horizon $r_d$ as a free parameter (marginalizing over it), the constraint on $H_0$ degrades as expected due to the loss of the absolute distance anchor.
Result: $H_0 = 70.9 \pm 2.9$ km s $^{-1}$ Mpc $^{-1}$ .
This result remains consistent with both Planck and local distance ladder measurements, demonstrating that DESI clustering alone can constrain $H_0$ without relying on early-universe physics, though with lower precision.

Evolving Dark Energy ( $w_0 w_a$ )

Combining DESI clustering with a compressed CMB likelihood yields constraints consistent with a cosmological constant ( $\Lambda$ ) within $2\sigma$ .
Results: $w_0 = -0.62 \pm 0.29$ , $w_a = -1.34 \pm 0.82$ .
The analysis finds no strong evidence for dynamical dark energy, though the central values show a slight preference for $w_0 > -1$ and $w_a < 0$ , similar to trends seen in other DESI analyses.

Neutrino Mass ( $\sum m_\nu$ )

DESI Only: The constraint is limited by the prior, $P(\sum m_\nu < 0.36 \text{ eV})$ at 95% CL.
DESI + CMB: Combining with Planck data tightens the constraint significantly, disfavoring the inverted hierarchy ( $\sum m_\nu \gtrsim 0.1$ eV) at $\sim 95\%$ confidence.
Result: $\sum m_\nu < 0.10$ eV (95% CL). This is consistent with the normal hierarchy ( $\sim 0.06$ eV).

5. Significance and Conclusion

This paper establishes the ShapeFit methodology as a powerful, model-independent tool for cosmological inference using high-precision large-scale structure data.

Efficiency vs. Precision: While the ShapeFit constraints on extended models (like $w_0 w_a$ or $\sum m_\nu$ ) are slightly weaker ( $\sim 10-30\%$ larger error bars) than full direct-shape modeling (due to information loss in compression and the lack of cross-covariance with reconstructed BAO in this specific implementation), the approach offers significant advantages in modularity and speed.
Bispectrum Utility: The inclusion of the bispectrum is shown to be critical for tightening constraints on structure growth parameters, providing a statistical gain equivalent to a massive increase in survey volume.
Future Outlook: The results validate the robustness of current DESI DR1 analyses. As DESI collects more data (increasing volume by an order of magnitude), the ShapeFit framework will serve as a foundational tool for testing new physics, provided that systematic effects (like window functions for the bispectrum) are further refined.

In summary, the work confirms that the standard $\Lambda$ CDM model remains robust against DESI DR1 data, with no significant deviations found, while demonstrating that the bispectrum is a vital component for maximizing the scientific return of future spectroscopic surveys.