DESI DR2 Favors Metastable in Emergent Dark Energy with No Evidence for Dark Matter–Dark Energy Interactions

Technical Summary: DESI DR2 Favors Metastable in Emergent Dark Energy with No Evidence for Dark Matter–Dark Energy Interactions

Problem Statement
The standard $\Lambda$CDM cosmological model, while empirically successful, faces significant theoretical challenges (the cosmological constant and coincidence problems) and growing observational tensions, specifically the Hubble ($H_0$) and structure growth ($S_8$) discrepancies. Recent data from the Dark Energy Spectroscopic Instrument Data Release 2 (DESI DR2), when combined with Cosmic Microwave Background (CMB) observations, has shown a preference for dynamical dark energy scenarios over a static cosmological constant, specifically exhibiting "Quintom-B" behavior ($w_0 > -1, w_a < 0, w_0 + w_a < -1$). Furthermore, the possibility of energy exchange between dark matter (DM) and dark energy (DE) remains a viable phenomenological extension to address these tensions. This work investigates the Interacting Metastable Emergent Dark Energy (IMEDE) model, which combines the late-time emergence of dark energy with a non-gravitational interaction between the dark sectors, to determine if it offers a statistically superior description of current cosmological data compared to $\Lambda$CDM.

Methodology
The authors performed a comprehensive Markov Chain Monte Carlo (MCMC) analysis using the Bayesian inference package Cobaya, coupled with the Boltzmann solver CAMB. The analysis constrained the IMEDE parameter space, defined as $\theta_{\text{IMEDE}} = \{\theta_{\Lambda\text{CDM}}, z_t, \Delta, Q\}$, where:

• $z_t$: The transition redshift where the equation of state changes most rapidly.
• $\Delta$: The amplitude of the transition from a phantom phase ($w < -1$) to a quintessence phase ($w > -1$).
• $Q$: A dimensionless coupling parameter governing the energy exchange between DM and DE (defined via the interaction term $Q = H Q \rho_{de}$).

The study utilized a combination of observational datasets:

1. CMB: Planck PR4 (NPIPE) temperature, polarization, and lensing likelihoods, along with ACT DR6 lensing data.
2. BAO: DESI DR2 baryon acoustic oscillation measurements across multiple tracers (BGS, LRG, ELG, QSO, Lyman-$\alpha$) spanning $0.3 \leq z \leq 2.33$.
3. Type Ia Supernovae (SNe Ia): Three independent compilations: Pantheon+, DES-Dovekie, and Union3.

The analysis included a stability check for the interaction parameter $Q$, restricting the prior to $-3 < Q < -0.001$ to avoid early-time gravitational instabilities (implying energy transfer from DM to DE). Model comparison was conducted using the logarithmic Bayes factor ($\ln B$) and the Gaussian significance ($N_\sigma$) derived from the difference in best-fit $\chi^2$ values ($\Delta \chi^2_{\text{MAP}}$).

Key Results

• Statistical Preference for IMEDE: The IMEDE model consistently provides a better fit to the data than $\Lambda$CDM.
  • Gaussian Significance: The preference ranges from weak to moderate. The CMB + DESI DR2 + Union3 combination yields the highest significance ($N_\sigma = 3.06$, $\Delta \chi^2_{\text{MAP}} = -14.58$).
  • Bayesian Evidence: The evidence is stronger. The CMB + DESI DR2 + Union3 combination provides decisive evidence ($\ln B = 10.33$) in favor of IMEDE, while other combinations show strong evidence.
• Origin of Preference: The statistical improvement is not driven by a single outlier data point. Instead, it arises from a coherent reduction of residuals across multiple independent DESI DR2 distance measurements (specifically radial distance $zH(z)$ and transverse comoving distance $D_M$) spanning the full redshift range. Notably, the tension at $z=0.70$ and $z=0.51$ is significantly reduced in the IMEDE model compared to $\Lambda$CDM.
• Dark Energy Dynamics: The reconstructed equation of state, $w(z)$, exhibits a Quintom-B evolution, transitioning from a phantom phase ($w < -1$) at earlier times to a quintessence phase ($w > -1$) at late times. The phantom divide crossing occurs at $z_t \simeq 0.3\text{--}0.5$.
• Dark Sector Interaction: The analysis finds no evidence for a non-zero interaction between dark matter and dark energy. The coupling parameter $Q$ is constrained to be consistent with zero, with lower bounds (e.g., $Q > -0.116$ at 95% C.L. for CMB + DESI DR2 + Union3) indicating that current data do not support a statistically significant interaction.
• Cosmological Tensions ($H_0$ and $S_8$): The IMEDE model does not resolve the $H_0$ or $S_8$ tensions.
  • Because the model only modifies late-time expansion history, the sound horizon at the drag epoch ($r_d$) remains unchanged ($\approx 147.6$ Mpc). Consequently, the inferred $H_0$ values ($65.2\text{--}67.5$ km s$^{-1}$ Mpc$^{-1}$) remain in significant tension with local SH0ES measurements ($73.04 \pm 1.04$).
  • Similarly, the predicted $S_8$ values ($0.83\text{--}0.89$) remain systematically higher than those preferred by weak-lensing surveys (e.g., DES Y6, KiDS-Legacy).

Significance and Claims
The paper claims that the IMEDE model represents a statistically well-motivated extension of the standard cosmological model. The primary significance lies in the model's ability to provide a more consistent description of the DESI DR2 BAO distance measurements through a dynamical, emergent dark energy scenario that naturally accommodates a phantom-crossing (Quintom-B) behavior.

The authors emphasize that the preference for IMEDE over $\Lambda$CDM is robust across different dataset combinations and is driven by a global improvement in fitting the distance ladder rather than a specific anomaly. However, the paper is modest regarding the resolution of cosmological tensions; it explicitly states that while the model improves the fit to expansion history data, it fails to alleviate the $H_0$ and $S_8$ tensions due to the preservation of early-universe physics. The authors conclude that these results motivate further testing with forthcoming DESI DR3 data and next-generation surveys (Euclid, Rubin Observatory, Roman Space Telescope) to confirm whether the observed preference for dynamical, emergent dark energy persists.