The cosmological constant Λ is a cornerstone of the standard model, yet its nature remains a mystery. Recent DESI DR2 data on baryon acoustic oscillations (BAO) point to deviations from ΛCDM, consistent with a growing body of work, including studies of dynamical dark energy and modified gravity. Our work continues this line by systematically comparing several dark energy parameterizations on combined datasets.
We performed a Bayesian analysis using the SimpleMC code, which employs the Metropolis-Hastings algorithm. The analysis includes BAO data from DESI DR2, three Type Ia supernova compilations (Pantheon+, DES-SN5Y, Union3), and a compressed CMB likelihood (PLK15). Models considered include ΛCDM, ωCDM, ω0ωaCDM, logarithmic, exponential, JBP, BA, and GEDE, as well as non-flat variants oΛCDM and oωCDM. For each model, the sum of neutrino masses and the effective number of relativistic species were varied.
All dynamical dark energy models favor the Quintom-B scenario: ω0 > −1, ωa < 0, and ω0 + ωa < −1. Deviations from ΛCDM reach 3σ when using DES-SN5Y or Union3, but do not exceed 5σ. Spatial curvature remains compatible with zero. Upper limits on the sum of neutrino masses range from <0.029 eV (GEDE with DES-SN5Y) to <0.520 eV (oωCDM), with ΛCDM giving a limit of <0.066 eV. The effective number of relativistic species is consistent with the standard value of 3.044. Bayesian factors show moderate to strong preference for dynamical models depending on the dataset.
This work strengthens the growing evidence for dark energy evolution, but does not provide a final verdict. Unlike many previous studies, we consider a wide range of parameterizations and account for neutrinos and curvature, allowing us to assess the robustness of the conclusions. The obtained limits on neutrino mass depend on the choice of dark energy model, which is important for interpreting future data.
If the deviations are confirmed with increasing precision from DESI and other surveys, this would require a revision of the standard model. Possible avenues include scalar fields, modified gravity, or interactions between dark components. It is also necessary to refine supernova systematics, especially at low redshifts.
The results will impact cosmology, neutrino physics, and the interpretation of data from major surveys such as DESI, Euclid, and Roman.
Further analysis with full DESI data, accounting for supernova systematics, and testing models against weak lensing data.
The key unresolved problem is the nature of dark energy. Our results point to dynamics but do not solve the cosmological constant problem. The question of neutrino mass also remains open: the limits depend on the dark energy model, which requires caution in interpretation.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
This work systematically compares several dark energy parameterizations (ωCDM, ω0ωaCDM, logarithmic, exponential, JBP, BA, GEDE, and non-flat variants) using combined DESI DR2 BAO data, three supernova compilations (Pantheon+, DES-SN5Y, Union3), and compressed CMB data (PLK15). Employing Bayesian analysis with the SimpleMC code, the authors find that all dynamical models favor the Quintom-B scenario (ω0 > −1, ωa < 0, ω0 + ωa < −1) with deviations up to 3σ, but not exceeding 5σ. They derive constraints on the sum of neutrino masses and the effective number of relativistic species, showing how these limits depend on the choice of dark energy model.
Where its strength is
The broad coverage of parameterizations, along with simultaneous variation of neutrino parameters and curvature, allows an assessment of the robustness of conclusions about dynamical dark energy and reveals the model dependence of neutrino mass constraints.
Where it could go next
Worth trying
Why this may matter
Confirming the evolution of dark energy, especially the crossing of the phantom divide, would be a revolution in cosmology, pointing to new physics beyond the standard model. Furthermore, model-dependent limits on neutrino mass are important for planning future experiments such as PICO or CMB-S4, which aim to measure the sum of masses with high precision.
Nearby works in our archive
The closest work in the archive sits at 0.67 by meaning — the usual level for our corpus.
If dark energy is indeed evolving, then in the past it might have behaved like a phantom (ω < −1), while today it behaves like quintessence (ω > −1). Such a transition is called phantom crossing, and it is forbidden in the simplest models.
In Liu Cixin's novel 'The Three-Body Problem', dark energy plays a key role in the fate of the universe, though its nature remains a mystery. Our results echo the idea that dark energy might not be constant but a dynamic quantity.