Searching for systematics in SNIa and galaxy cluster data using the cosmic duality relation

Arman Shafieloo, Subhabrata Majumdar, Varun Sahni, Alexei A. StarobinskyJournal of Cosmology and Astroparticle Physics 2013 · DOI · arXiv

Abstract

We compare two different probes of the expansion history of the universe: luminosity distances from Type Ia supernovae (SNIa) and angular diameter distances from galaxy clusters. By employing a Bayesian interpretation of the Crossing statistic in conjunction with the Cosmic Duality Relation (CDR), we develop a model-independent method to identify inconsistencies between these datasets. Such inconsistencies would imply the presence of systematic errors in either the SNIa or galaxy cluster data. Using simulated observations based on expected WFIRST (Nancy Grace Roman Space Telescope) supernovae data and X-ray eROSITA + SZ Planck cluster data, we demonstrate that this method can detect systematics with high precision without requiring a-priori assumptions about the nature of dark energy.

Introduction

In the current era of precision cosmology, understanding the universe's accelerated expansion depends on the quality of our observational data. Two primary distance measures are used to reconstruct expansion history: luminosity distances ($d_L$) from SNIa and angular diameter distances ($d_A$) from galaxy clusters. However, both are prone to systematic errors—unintended biases in measurement or modeling.

Traditionally, researchers detect these errors by assuming a specific cosmological model, such as $\Lambda$CDM. However, if the underlying model is incorrect, the detected "errors" may simply be a failure of the model itself. This paper proposes a method to identify data inconsistencies independently of any assumptions regarding dark energy or gravity.

The Cosmic Duality Relation as a Diagnostic Tool

The core of this approach is the Cosmic Duality Relation (CDR), a robust geometric principle stating that in a standard spacetime where photons are conserved, the two distance measures are related by:

$$d_L = (1 + z)^2 d_A$$

where $z$ is the redshift. While the CDR is often used to search for exotic physics (like photon-axion conversion), this study treats it as an exact principle to identify observational systematics. If the relation is violated, it indicates a discrepancy between the datasets that must be addressed.

Methodology

The authors combine two sophisticated statistical tools to perform a model-independent comparison:

  1. The Smoothing Method: A non-parametric technique used to extract a mean function from noisy data without assuming a specific formula for the expansion history.
  2. Crossing Statistics: A Bayesian approach using Chebyshev polynomials to identify "crossings" or deviations between datasets. By mapping these into a hyperparameter space ($C_1, C_2$), the method creates confidence contours. Overlapping contours indicate consistency, while divergence reveals systematic errors.

Analysis of Simulated Data

To test the method's sensitivity for future surveys, the authors utilized simulated catalogs:

  • SNIa (WFIRST-like): Approximately 2,300 supernovae ranging from $z = 0.015$ to $1.7$.
  • Galaxy Clusters (eROSITA + Planck-like): Between 1,830 and 2,000 clusters ranging from $z = 0.1$ to $1.1$.

These were simulated using a standard $\Lambda$CDM model as a baseline, but the method was also tested against a "quiescence" model ($w = -0.8$) to confirm its model-independence.

Key Results and Insights

  • High Sensitivity: The method successfully detected systematic shifts as small as $\alpha = 0.002$ in the distance modulus, representing sub-percent level errors that are difficult to find using traditional methods. Even when datasets appear consistent to the naked eye on standard distance plots, the Crossing Statistic can clearly separate them in hyperparameter space.
  • Robustness to $H_0$ Uncertainty: Uncertainty in the Hubble constant ($H_0$) expands error contours in a direction perpendicular to the overlap, meaning it does not obscure the signature of systematic errors.
  • Model Independence: The test performed equally well regardless of whether the underlying universe followed $\Lambda$CDM or an alternative dark energy model. This ensures that the detection of a systematic is not a byproduct of an incorrect cosmological assumption.

Conclusion

This model-independent approach offers a powerful way to "fact-check" cosmological observations. By leveraging the CDR and Crossing Statistics, researchers can identify sub-percent level inconsistencies in future datasets from missions like the Nancy Grace Roman Space Telescope and eROSITA. Identifying these discrepancies is a vital first step in ensuring the accuracy of our understanding of the universe's expansion.

Summary updated

Abstract

We compare two different probes of the expansion history of the universe, namely, luminosity distances from type Ia supernovae and angular diameter distances from galaxy clusters, using the Bayesian interpretation of Crossing statistic [1, 2] in conjunction with the assumption of cosmic duality relation. Our analysis is conducted independently of any a-priori assumptions about the nature of dark energy. The model independent method which we invoke searches for inconsistencies between SNIa and galaxy cluster data sets. If detected such an inconsistency would imply the presence of systematics in either of the two data sets. Simulating observations based on expected WFIRST supernovae data and X-ray eROSITA + SZ Planck cluster data, we show that our method allows one to detect systematics with high precision and without advancing any hypothesis about the nature of dark energy.
BibTeX
@article{shafieloo2013searching,
  author = {Arman Shafieloo and Subhabrata Majumdar and Varun Sahni and Alexei A. Starobinsky},
  title = {{Searching for systematics in SNIa and galaxy cluster data using the cosmic duality relation}},
  journal = {Journal of Cosmology and Astroparticle Physics},
  year = {2013},
  doi = {10.1088/1475-7516/2013/04/042},
}
Figure