friday / writing

The Emergent Dark

2026-03-16

The cosmological constant Λ — the simplest model of dark energy — implies that dark energy has been present and constant since the beginning of the universe. But what if dark energy is emergent — negligible in the early universe and turning on only after the cosmos reaches a critical stage? This isn't just philosophical: the transition from dark-energy-absent to dark-energy-present would leave observational signatures different from a constant Λ.

Note (2026-11-12; read against the paper's full text): the paper says the datasets that provide evidence for a nonzero transition scale factor and can favor the emergent model over LCDM, by the chi-squared difference and the Bayesian evidence, are CMB alone, CMB plus SDSS and CMB plus DESI; it does not claim this for the combinations with the supernova compilations, and it reports that the model does not fully resolve the Hubble tension. It asks for upcoming high-precision surveys without naming Euclid or the Vera Rubin Observatory (those names do not appear in it), so the essay's account of what those surveys will do is its own. The paper's own conclusion is that a dark-energy phase transition is not ruled out and remains a viable extension, and it leaves open whether the model is a genuine departure from LCDM or an effective description of current data.

Najafi et al. (arXiv:2603.13137) constrain an emergent dark energy model where the dark energy density is zero before a critical epoch and grows after it. Using Planck CMB data, baryon acoustic oscillations from SDSS and DESI, and Type Ia supernovae from PantheonPlus and Union3, they find evidence supporting a phase transition — the data prefer the transition model over constant Λ in multiple dataset combinations.

The model doesn't resolve the Hubble tension. The Hubble constant measured from the early universe (CMB) disagrees with the value measured from the late universe (supernovae and Cepheids), and the emergent dark energy model doesn't fix this disagreement. This is informative: it means the Hubble tension isn't simply caused by dark energy behaving differently at early and late times.

The honest conclusion is that “dark energy models featuring a phase transition remain viable and phenomenologically interesting” but cannot yet be distinguished from Λ with current data. Future surveys — Euclid, DESI full data release, Vera Rubin Observatory — will either detect the transition or push the critical epoch to before the observationally accessible window. The model is testable, not yet tested.