The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: Examining the observational evidence for dynamical dark energy

Gong‐Bo Zhao, Marco Raveri, Levon Pogosian, Yuting Wang, Robert Crittenden, Will Handley, Will J. Percival, J. Brinkmann, Chia-Hsun Chuang, Antonio J. Cuesta, Daniel J. Eisenstein, Francisco-Shu Kitaura, K. Koyama, Benjamin L’Huillier, R. C. Nichol, Matthew M. Pieri, Sergio Rodríguez-Torres, Ashley J. Ross, Graziano Rossi, Ariel G. Sánchez, Arman Shafieloo, Jeremy L. Tinker, Rita Tojeiro, J. Alberto Vázquez, Hanyu Zhang

Abstract

A flat Friedman-Roberson-Walker universe dominated by a cosmological constant ($\Lambda$) and cold dark matter (CDM) has served as the standard model of cosmology since the discovery of cosmic acceleration. However, as observational precision increases, tensions have emerged between different datasets within the $\Lambda$CDM framework. This paper investigates whether these discrepancies—specifically regarding the Hubble constant ($H_0$) and matter density ($\Omega_M$)—can be resolved by transitioning from a constant dark energy model to a dynamical one. Using data from the completed SDSS-III Baryon Oscillation Spectroscopic Survey (BOSS), we find that a dynamical dark energy model is preferred at a 3.5$\sigma$ significance level, effectively relieving existing tensions. While current Bayesian evidence remains inconclusive, we demonstrate that the upcoming DESI survey will have the sensitivity to definitively confirm or rule out this evolution.

Introduction: The Snag in the Standard Model

For nearly two decades, the $\Lambda$CDM model has been the "Gold Standard" of cosmology. It operates on the assumption that dark energy is a cosmological constant ($\Lambda$), meaning its energy density remains uniform as the universe expands. However, recent measurements of the early and late universe have begun to diverge, creating two primary areas of concern:

  1. The $H_0$ Tension: Local measurements of the universe's expansion rate ($H_0$) using supernovae suggest a faster expansion ($73.24 \pm 1.74$ km s$^{-1}$ Mpc$^{-1}$) than the value inferred from the Planck Cosmic Microwave Background (CMB) data ($66.93 \pm 0.62$ km s$^{-1}$ Mpc$^{-1}$). This represents a $3.4\sigma$ discrepancy.
  2. The $\Omega_M$ Tension: Measurements of Baryon Acoustic Oscillations (BAO) in the "Lyman-$\alpha$ forest"—distant gas clouds—prefer a lower matter density fraction ($\Omega_M$) than the value preferred by the CMB.

These tensions raise a fundamental question: Are these discrepancies the result of systematic measurement errors, or is the $\Lambda$CDM model itself incomplete? This study explores the possibility that dark energy is not a constant, but rather evolves over cosmic time.

Methodology: Quantifying Surprise

To test the validity of dynamical dark energy (DE), the authors utilized a sophisticated statistical toolkit designed to compare models without bias:

  • KL Divergence and "Surprise": The researchers used Kullback-Leibler (KL) divergence to quantify the tension between datasets. They calculated a metric termed "Surprise"—the difference between the actual and expected information gain—to determine if two datasets are truly inconsistent within a specific model.
  • Non-parametric Reconstruction: Rather than assuming a fixed mathematical formula for dark energy's evolution ($w(z)$), the authors employed a "correlated prior" method. They divided the history of the universe into 30 bins, allowing the data to dictate the shape of the evolution curve, requiring only that transitions between bins remain relatively smooth.
  • Bayesian Evidence: To ensure the model's complexity was justified, they calculated Bayes factors. This approach penalizes models for adding extra parameters (like the 30 bins of DE evolution) unless they provide a significantly better fit to the data.

Data Synthesis

The study relies on a comprehensive combined dataset referred to as ALL16, which integrates:

  • CMB: Planck 2015 temperature and polarization data.
  • Supernovae (SNe): The Joint Light-curve Analysis (JLA) sample.
  • BAO: Measurements from the BOSS DR12, including the critical Lyman-$\alpha$ forest data.
  • $H_0$: The 2016 local Hubble constant measurement from Cepheid variables.
  • OHD: "Cosmic Chronometers" measuring the relative ages of old galaxies.

To track the evolution of evidence, these results were compared against ALL12, a smaller dataset from 2012, illustrating how the statistical significance of these findings has grown over four years.

Key Insights: The Phantom Crossing

One of the most physically significant findings of the reconstruction is that the dark energy equation of state ($w$) appears to cross the "-1 boundary." In standard $\Lambda$CDM, $w$ is exactly -1. A model where $w$ crosses this boundary is known as a "phantom" crossing. This is a crucial discovery because simple "quintessence" models (involving a single scalar field) cannot produce this behavior. It suggests that if dark energy is indeed dynamical, it may require more exotic physics, such as multiple interacting fields or Modified Gravity.

Furthermore, the dynamical model provides massive tension relief. Under $\Lambda$CDM, the tension between $H_0$ and other datasets sits at a significant 4.4$\sigma$. When the model allows for dynamical dark energy, this tension effectively vanishes, dropping to a negligible 0.7$\sigma$.

Results and Future Outlook

The dynamical dark energy model is preferred over $\Lambda$CDM at a 3.5$\sigma$ significance level based on the improvement in the fit (a $\chi^2$ improvement of -12.3). However, the Bayesian evidence—which accounts for model complexity—does not yet "favor" the dynamical model decisively. The Bayes factor currently remains near zero, suggesting the data is "on the fence."

While the observed tensions could still potentially be attributed to unknown systematic errors in the data (such as calibration issues in supernovae or gas cloud measurements), the consistency of the signal between 2012 and 2016 data is striking.

The resolution of this cosmological standoff lies with the next generation of "Stage IV" dark energy experiments. The authors simulated the upcoming DESI (Dark Energy Spectroscopic Instrument) survey and found that if the current best-fit dynamical model is correct, DESI will provide "decisive" Bayesian evidence (a Bayes factor of ~11.3) to confirm it, finally moving beyond the cosmological constant.

Summary updated

Abstract

A flat Friedman-Roberson-Walker universe dominated by a cosmological constant ($\Lambda$) and cold dark matter (CDM) has been the working model preferred by cosmologists since the discovery of cosmic acceleration. However, tensions of various degrees of significance are known to be present among existing datasets within the $\Lambda$CDM framework. In particular, the Lyman-$\alpha$ forest measurement of the Baryon Acoustic Oscillations (BAO) by the Baryon Oscillation Spectroscopic Survey (BOSS) prefers a smaller value of the matter density fraction $\Omega_{\rm M}$ compared to the value preferred by cosmic microwave background (CMB). Also, the recently measured value of the Hubble constant, $H_0=73.24\pm1.74 {\rm km} {\rm s}^{-1} {\rm Mpc}^{-1}$, is $3.4\sigma$ higher than $66.93\pm0.62 {\rm km} {\rm s}^{-1} {\rm Mpc}^{-1}$ inferred from the Planck CMB data. In this work, we investigate if these tensions can be interpreted as evidence for a non-constant dynamical dark energy (DE). Using the Kullback-Leibler (KL) divergence to quantify the tension between datasets, we find that the tensions are relieved by an evolving DE, with the dynamical DE model preferred at a $3.5\sigma$ significance level based on the improvement in the fit alone. While, at present, the Bayesian evidence for the dynamical DE is insufficient to favour it over $\Lambda$CDM, we show that, if the current best fit DE happened to be the true model, it would be decisively detected by the upcoming DESI survey.
BibTeX
@inproceedings{zhao2017clustering,
  author = {Gong‐Bo Zhao and Marco Raveri and Levon Pogosian and Yuting Wang and Robert Crittenden and Will Handley and Will J. Percival and J. Brinkmann and Chia-Hsun Chuang and Antonio J. Cuesta and Daniel J. Eisenstein and Francisco-Shu Kitaura and K. Koyama and Benjamin L’Huillier and R. C. Nichol and Matthew M. Pieri and Sergio Rodríguez-Torres and Ashley J. Ross and Graziano Rossi and Ariel G. Sánchez and Arman Shafieloo and Jeremy L. Tinker and Rita Tojeiro and J. Alberto Vázquez and Hanyu Zhang},
  title = {{The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: Examining the observational evidence for dynamical dark energy}},
  booktitle = {Nature Astronomy},
  year = {2017},
  doi = {10.1038/s41550-017-0216-z},
}
Figure