Reanalyzing DESI DR1: 2. Constraints on Dark Energy, Spatial Curvature, and Neutrino Masses

This paper presents an independent re-analysis of DESI DR1 data using full-shape power spectrum and bispectrum measurements to derive robust, CMB-independent constraints on spatial curvature, dark energy dynamics, and neutrino masses, demonstrating significant improvements in cosmological parameter limits compared to standard BAO-only analyses.

The Gist

The Big Picture: The Cosmic Detective Story

Imagine the universe as a giant, expanding balloon. For decades, scientists have been trying to figure out exactly how fast this balloon is inflating, what is inside it, and what is pushing it to expand faster.

The standard theory, called $\Lambda$CDM, is like a well-worn map. It says the universe is made of normal stuff, invisible "dark matter," and a mysterious force called "dark energy" that acts like a constant push. This map has worked great for a long time.

However, a new, high-tech telescope called DESI (Dark Energy Spectroscopic Instrument) has started taking incredibly detailed photos of the universe. When scientists looked at these new photos, they noticed some things didn't quite match the old map. The universe seemed to be expanding a bit differently than expected, and the "push" (dark energy) might be changing over time.

This paper is like a team of expert mechanics (the authors) taking the DESI data, putting it through a brand-new, super-precise diagnostic tool, and checking if the "engine" of the universe needs a tune-up or a completely new part.

---

The New Diagnostic Tool: "Full-Shape" Analysis

In the past, scientists looked at the DESI data like a tourist looking at a mountain range from far away. They saw the big peaks (called Baryon Acoustic Oscillations or BAO) and used those to measure distances. It's like measuring a mountain's height by looking at its shadow.

In this paper, the authors use a method called Full-Shape (FS) analysis.

• The Analogy: Instead of just looking at the shadow, they are now climbing the mountain and feeling every rock, ridge, and valley. They are analyzing the entire shape of the data, not just the big peaks.
• The Secret Weapon: They also look at the Bispectrum. If the Power Spectrum is a photo of the universe, the Bispectrum is a 3D movie that shows how the galaxies are interacting and clustering together in complex patterns. It's like listening to the sound of the universe, not just looking at the picture.

By using this "Full-Shape" tool, they can hear and see details that the old "shadow" method missed.

---

The Three Big Mysteries They Solved

The team used this new tool to test three specific theories about what might be wrong with our cosmic map.

1. Is the Universe Curved? (Spatial Curvature)

• The Question: Is the universe flat like a sheet of paper, or curved like a saddle (open) or a ball (closed)?
• The Old View: The old map said it was perfectly flat.
• The New Finding: When they added the "Full-Shape" data, they found the universe is still flat, but they are now twice as sure about it.
• The Analogy: Imagine trying to guess if a table is flat by looking at it from one angle. You might be 50% sure. Now, imagine walking around the table, touching every corner, and measuring it with a laser. You are now 100% sure it's flat. The new data doubled their confidence.

2. Is Dark Energy Changing? (Dynamical Dark Energy)

• The Question: Is the "push" of dark energy a constant force (like a battery with steady voltage), or is it a variable force (like a dimmer switch that changes brightness)?
• The Old View: The standard map says it's a constant battery ($\Lambda$).
• The New Finding: The data hints that the "dimmer switch" might be on. The "push" seems to be changing slightly over time.
• The Analogy: Think of the universe's expansion like a car accelerating. The old map said the driver was pressing the gas pedal at a steady, fixed rate. The new data suggests the driver might be gently easing off or pressing harder depending on the road.
• The Result: Adding the new "Full-Shape" data made the evidence for this "changing driver" about 30% stronger (without supernova data) and 20% stronger (with supernova data). It's not a slam-dunk proof yet, but the hint is getting louder.

3. How Heavy are Neutrinos? (Neutrino Mass)

• The Question: Neutrinos are ghostly particles that fly through everything. We know they have mass, but how much?
• The Old View: Cosmology usually says they are very light.
• The New Finding: This paper set the strictest limit yet on how heavy they can be, without relying on the Cosmic Microwave Background (CMB) data.
• The Analogy: Imagine trying to weigh a feather by seeing how much it slows down a speeding car. The new "Full-Shape" data is like adding a high-speed camera to the car. They found that the feather (neutrinos) must be incredibly light—lighter than a specific tiny threshold.
• The Result: They found that if the universe is the standard kind, neutrinos must weigh less than 0.059 eV. If the universe is curved or dark energy is changing, the limit gets a bit looser, but it's still the tightest constraint we have.

---

Why This Matters: The "Ghost" in the Machine

The most exciting part of this paper isn't just the numbers; it's the method.

For a long time, scientists were afraid that if they tried to measure these complex things (like changing dark energy) without the "supernova" data (which has its own problems), they would get confused by "ghosts" in the data—mathematical errors that make it look like a change is happening when it isn't.

The authors developed a new way to clean up the data (using better "priors" and re-parameterization).

• The Analogy: It's like they invented a new noise-canceling headphone for the universe. Before, the static was so loud you couldn't hear the music. Now, they can hear the music clearly, even without the "supernova" backup singers.

The Bottom Line

This paper tells us:

1. The Universe is likely flat, and we are much more confident about it.
2. Dark Energy might be changing, and the new "Full-Shape" data makes this possibility much more likely than before.
3. Neutrinos are very light, and we have a new, very strict rule for how heavy they can be.
4. The New Tool Works: By looking at the entire shape of the galaxy distribution (not just the peaks), we can learn more about the universe's secrets without needing to rely on other, potentially messy data sources.

In short, the authors took a blurry photo of the universe, sharpened it with a new lens, and found that the universe is a bit more dynamic and mysterious than we thought, but also a bit more stable in its shape than we feared.

Technical Summary

1. Problem Statement

The standard cosmological model ($\Lambda$CDM) faces several theoretical and observational challenges. Recent data from the Dark Energy Spectroscopic Instrument (DESI) and the Cosmic Microwave Background (CMB) have hinted at potential deviations from the standard model, including:

• Dynamical Dark Energy (DDE): Tensions suggesting the dark energy equation of state ($w$) may not be constant ($w \neq -1$), with some analyses favoring "phantom-crossing" behavior ($w < -1$ evolving to $w > -1$).
• Spatial Curvature: Indications of non-zero spatial curvature ($\Omega_k \neq 0$).
• Neutrino Masses: The sum of neutrino masses ($M_\nu$) is a free parameter in the Standard Model, yet cosmological bounds often conflict with particle physics expectations (e.g., inverted hierarchy) depending on the assumed cosmological model.
• Systematics and Degeneracies: Previous analyses often relied heavily on Supernovae (SNe) data to break degeneracies in extended models, introducing potential systematic uncertainties. Furthermore, "prior volume effects" in Bayesian analyses of non-minimal models can bias parameter recovery.

The authors aim to perform an independent re-analysis of the DESI Data Release 1 (DR1) to constrain these non-minimal extensions ($\Lambda$CDM+$M_\nu$, $o\Lambda$CDM, $w_0w_a$CDM) using Full-Shape (FS) clustering information (power spectrum and bispectrum) combined with CMB and BAO data, without relying solely on SNe.

2. Methodology

The authors utilize an independent analysis pipeline based on the Effective Field Theory of Large Scale Structure (EFTofLSS), building upon their previous work (Paper 1).

• Datasets:

  • DESI DR1: Full-shape power spectrum ($P_\ell$) and bispectrum ($B_0$) measurements from six galaxy/quasar samples (BGS, LRG1-3, ELG2, QSO) spanning $z \in [0.1, 2.1]$.
  • DESI DR2: Official Baryon Acoustic Oscillation (BAO) parameters.
  • CMB: Planck 2018 (plik) and newer Planck PR4 (HiLLiPoP+LoLLiPoP) likelihoods, including lensing.
  • SNe: Three compilations (Pantheon+, Union3, DES-SN5YR) used for comparison but not strictly required for the primary FS constraints.

• Theoretical Framework:

  • Power Spectrum: Modeled at one-loop order in EFT, including monopole, quadrupole, and hexadecapole moments ($\ell=0, 2, 4$) up to $k_{max} = 0.20 \, h\text{Mpc}^{-1}$.
  • Bispectrum: Modeled at tree-level (large scales only), utilizing the monopole moment up to $k_{max} = 0.08 \, h\text{Mpc}^{-1}$.
  • Nuisance Parameters: The model includes 14 nuisance parameters per redshift chunk (bias parameters, stochastic counterterms, etc.).
  • Reparameterization: A key methodological improvement involves rescaling nuisance parameters by the Alcock-Paczynski (AP) amplitude and $\sigma_8(z)$. This mitigates "prior volume" projection effects that previously biased results in extended cosmological models.
  • Neutrinos: Modeled using the Einstein-de Sitter (EdS) approximation with a degenerate mass spectrum. Scale-dependent bias effects are neglected as they are sub-dominant to current measurement precision.

• Analysis:

  • Markov Chain Monte Carlo (MCMC) sampling using Montepython.
  • Marginalization over nuisance parameters is performed analytically where possible to reduce computational cost and bias.
  • Convergence is checked using the Gelman-Rubin criterion ($R-1 < 0.02$).

3. Key Contributions

1. First CMB-Independent Dark Energy Constraints: By utilizing the FS power spectrum and bispectrum with improved EFT priors, the authors obtain robust constraints on dynamical dark energy ($w_0, w_a$) from CMB + DESI (BAO+FS) data without relying on Supernovae distance measurements.
2. Inclusion of Bispectrum and Hexadecapole: This is the first application of the DESI DR1 bispectrum monopole and power spectrum hexadecapole in a full-shape analysis of non-minimal models, demonstrating their utility in breaking degeneracies.
3. Mitigation of Prior Volume Effects: The paper validates a new set of EFT parameter priors (AP-rescaling) that significantly reduces bias in parameter recovery for $w_0w_a$CDM models, addressing a known issue in previous DESI analyses.
4. Independent Verification: Provides an independent cross-check of the official DESI collaboration results using a distinct pipeline and theoretical assumptions.

4. Key Results

A. Spatial Curvature ($\Omega_k$)

• DESI-Only: Adding FS data to BAO improves constraints on $\Omega_k$ by a factor of 2 compared to BAO alone. The result is consistent with a flat universe ($\Omega_k = 0$).
• Combined (CMB+DESI): The addition of FS data to CMB+BAO does not significantly improve curvature constraints compared to CMB+BAO alone, as CMB already provides tight curvature limits.
• Significance: The combined data shows a mild preference for positive curvature ($\Omega_k > 0$) at $\sim 2.3\sigma$, but this vanishes when using the newer Planck PR4 likelihoods, remaining consistent with a flat universe.

B. Dynamical Dark Energy ($w_0w_a$CDM)

• Figure of Merit (FoM): The addition of DESI DR1 FS data increases the FoM for $w_0$ and $w_a$ by ~30% (without SNe) and ~20% (with SNe) relative to CMB+BAO results.
• Tension with $\Lambda$CDM: The preference for dynamical dark energy ($w_0 > -1, w_a < 0$) remains, with significances ranging from 2.9$\sigma$ to 4.3$\sigma$ depending on the dataset combination (highest with DESY5 SNe).
• Independence from SNe: Crucially, the FS data allows for robust constraints on $w_0, w_a$ without SNe, confirming that the preference for DDE is not an artifact of supernova systematics.
• $\Omega_m$ Shift: The inclusion of FS data shifts the inferred matter density $\Omega_m$ lower, bringing it into better agreement with DESI-only $\Lambda$CDM results.

C. Neutrino Masses ($M_\nu$)

• CMB-Independent Bounds: The analysis yields the strongest CMB-independent constraint on total neutrino mass: $M_\nu < 0.32$ eV (95% CL). The bispectrum contributes a 30% improvement over power spectrum-only results.
• Combined Bounds:
  • $\Lambda$CDM+$M_\nu$: $M_\nu < 0.059$ eV (14% improvement over CMB+BAO). This excludes the inverted hierarchy at $3.0\sigma$.
  • $o\Lambda$CDM+$M_\nu$: $M_\nu < 0.097$ eV (22% improvement).
  • $w_0w_a$CDM+$M_\nu$: $M_\nu < 0.13$ eV (22% improvement).
• Significance: Allowing for non-minimal backgrounds (curvature or dynamical DE) relaxes the neutrino mass bounds, preventing the exclusion of the inverted hierarchy in these extended scenarios.

5. Significance and Implications

• Robustness of Extended Models: The paper demonstrates that robust constraints on non-minimal cosmological models can be achieved using LSS full-shape data without relying on potentially systematics-prone supernova data.
• Validation of EFT Priors: The study confirms that reparameterizing EFT nuisance parameters is essential for unbiased inference in extended cosmologies, resolving previous tensions regarding prior volume effects.
• Neutrino Physics: The results highlight the sensitivity of neutrino mass constraints to the assumed cosmological background. While $\Lambda$CDM strongly disfavors the inverted hierarchy, allowing for dynamical dark energy or curvature restores compatibility with it.
• Future Prospects: The methodology sets the stage for analyzing future DESI data releases (DR2 and beyond) and other large-scale structure surveys, particularly in testing the nature of dark energy and the fundamental properties of neutrinos.

In conclusion, this work establishes that incorporating full-shape power spectrum and bispectrum measurements significantly enhances the constraining power of cosmological data, providing tighter, more robust limits on dark energy dynamics, spatial curvature, and neutrino masses, while offering a critical independent check on the standard cosmological paradigm.