Constranits of dynamical dark energy models from different observational datasets
Peiyuan Xu, Lu Chen, Guohao Li, Yang Han
TL;DR
This study systematically constrains seven dynamical dark energy parameterizations (including $w$CDM, CPL, JBP, FSLL, BA, LOG, EXP) against three up-to-date dataset combinations that pair CMB measurements (Planck PR4) with BAO (DESI DR2) and Type Ia supernova compilations (PantheonPlus, DES-Y5, Union3). Across all models, the inferred $H_{0}$ values are consistently lower than in $\Lambda$CDM, whereas $\sigma_{8}$ is reduced, suggesting limited relief of the $H_{0}$ tension but improved handling of $\sigma_{8}$ tension. The DES-Y5 dataset notably improves convergence to unimodal posteriors for $w_{0}$ and enhances model-fit quality for several two-parameter DDE forms, with the Barboza-Alcaniz (BA), Logarithmic (LOG), and Exponential (EXP) parameterizations often outperforming $\Lambda$CDM in terms of the AIC, especially BA. The results demonstrate strong dataset dependence in constraining dynamical dark energy and point to BA as a particularly promising two-parameter model under current observational data, motivating further theoretical exploration of its physical implications.
Abstract
The measurements of baryon acoustic oscillation by the Dark Energy Spectroscopic Instrument Data Release 2 indicate that dark energy may be a dynamical quantity with a time-varying equation of state. This challenges the core assumptions of the $Λ$CDM model and has generated significant interest in dynamical dark energy models. Therefore, studying the parameterization of the equation of state for dynamical dark energy is crucial. Existing work has achieved fruitful results in the dark energy models, exploring various parameterization forms, but it is relatively scattered and lacks systematic parameter constraints based on the latest dataset combinations. We use the $Λ$CDM as a baseline model and carry out rigorous statistical constraints on key cosmological parameters for seven representative parameterization models. Planck PR4 and DESI DR2 observations are incorporated into our study. We use three dataset combinations: CMB+BAO+PantheonPlus, CMB+BAO+DES-Y5, and CMB+BAO+Union3. The ${H}_{0}$ and ${σ}_{8}$ values of all dynamical dark energy models are lower than the $Λ$CDM model, indicating that our results may not effectively alleviate ${H}_{0}$ tension, but can significantly reduce ${σ}_{8}$ tension. By comparing the $χ^2$ and the Akaike Information Criterion obtained for each model, we demonstrate that the linear Chevallier-Polarski-Linder parameterization model is not the optimal choice in all cases. Specifically, when combined with the CMB+BAO+DES-Y5 dataset, the Barboza-Alcaniz, Logarithmic, and Exponential models demonstrate superior statistical fitting performance compared to the $Λ$CDM model. The Barboza-Alcaniz model shows a great advantage in fitting performance, leading to the most significant improvement.