Observational constraints on early time non-phantom behaviour of dynamical dark energy

This paper demonstrates that while late-time dynamical dark energy models can mildly improve cosmological fits, imposing early-time non-phantom scaling behavior is strongly disfavored by current observational data, as it requires steep potentials that fail to alleviate the Hubble tension and are penalized by Bayesian model selection.

The Gist

Imagine the universe as a giant, expanding balloon. For a long time, scientists thought this balloon was being inflated by a steady, unchanging force called the "Cosmological Constant" (like a perfectly steady hand pushing the balloon). This model, called ΛCDM, has been the gold standard for decades.

However, recent measurements of how fast the universe is expanding (the Hubble Tension) and how galaxies are clumping together have started to show cracks in this steady-hand theory. New data suggests the "push" might actually be changing over time, like a hand that speeds up or slows down. This changing force is called Dynamical Dark Energy.

This paper investigates a specific idea: What if this changing force acted differently in the past compared to today?

The Story of the "Double-Acting" Force

The authors propose a model where Dark Energy is like a two-stage rocket or a chameleon:

1. The Early Days (The "Scaling" Phase): In the early universe, when the cosmos was hot and dense with radiation and matter, this Dark Energy was supposed to be a "good citizen." It was supposed to scale with the background, meaning it grew at the same rate as everything else but stayed very small, never taking over the party. Think of it as a quiet guest at a loud concert who stays in the corner and doesn't disturb the music.
2. The Late Days (The "Thawing" Phase): As the universe expanded and cooled, the "Hubble friction" (a cosmic drag that kept the field frozen) weakened. The Dark Energy was supposed to "thaw" out, start moving, and begin pushing the universe apart faster, causing the acceleration we see today.

The authors tested several variations of this "two-stage" story using the latest data from the Planck satellite (looking at the baby universe), DESI (mapping galaxies), and Supernovae (exploding stars as distance markers).

The Main Findings: A Tale of Two Tensions

Here is what the data told them, using simple analogies:

1. The "Good Citizen" Rule is Too Strict
The data demanded that this early "scaling" behavior be extremely strict. To keep Dark Energy from taking over the early universe, the "steepness" of its potential energy had to be incredibly high (a value called $\lambda$ greater than 20 or 30).

• The Analogy: Imagine trying to keep a small dog (Dark Energy) from running away in a giant field (the early universe). The data says the leash has to be pulled so tight that the dog is practically frozen in place.
• The Result: Because the leash is so tight, the dog (Dark Energy) is forced to be so small in the early universe that it contributes less than 1% of the total energy. This means it cannot be the reason for the current "Hubble Tension." It's too weak to have changed the starting conditions of the universe enough to fix the speed discrepancy.

2. The "Changing Hand" is Still Popular (But Complicated)
When the authors looked only at the recent universe (late times), the data did show a slight preference for a changing Dark Energy, specifically one that behaves slightly "phantom-like" (pushing harder than a standard constant).

• The Analogy: It's like the data is saying, "The hand pushing the balloon isn't perfectly steady; it's getting a little stronger."
• The Catch: However, when you force this "changing hand" to also follow the strict "good citizen" rules of the early universe (the scaling phase), the model becomes too complicated.

3. The "Complexity Penalty"
The authors used statistical tools (AIC and BIC) to judge these models. Think of these tools as a judge who says, "If you want to add more moving parts to your story, you need to prove it's worth the extra complexity."

• The Verdict: The models that included the early "scaling" phase were penalized. Even though they fit the data slightly better in some ways, the extra complexity of forcing the early universe to behave so strictly wasn't worth it. The simplest model (the standard ΛCDM) still wins the award for the best balance of simplicity and accuracy.

The Bottom Line

The paper concludes that while the universe might be driven by a changing Dark Energy today, the idea that this energy followed a strict "scaling" path in the early universe is strongly disfavored by current data.

• Did it fix the Hubble Tension? No. The strict early rules kept the Dark Energy too small to make a difference.
• Is the standard model dead? Not quite. The data still slightly prefers a changing Dark Energy at late times, but forcing it to have a specific "early life" makes the theory too clunky to be the best explanation.

In short: The universe might be accelerating due to a changing force, but that force probably didn't play by the strict "scaling" rules the authors tested in the early days. The simplest explanation remains the most robust one.

Technical Summary

Technical Summary: Observational Constraints on Early Time Non-Phantom Behaviour of Dynamical Dark Energy

Problem Statement
The standard $\Lambda$CDM cosmological model faces significant challenges, most notably the $\sim 5\sigma$ tension between local measurements of the Hubble constant ($H_0$) by the SH0ES collaboration and values inferred from Cosmic Microwave Background (CMB) data under the $\Lambda$CDM framework. Additionally, the model struggles with the $S_8$ tension regarding matter clustering amplitude. Recent data releases from the Dark Energy Spectroscopic Instrument (DESI) suggest a preference for dynamical dark energy (DDE) over a pure cosmological constant, with statistical significance increasing from $2.6\sigma$ (DR1) to $3.1\sigma$ (DR2) when combined with CMB data.

While simple parametrizations like the Chevallier–Polarski–Linder (CPL) model can describe late-time acceleration, they often fail at high redshifts. To address early-time behavior, scalar field models with steep exponential potentials are often invoked to produce "scaling" solutions, where the scalar field energy density tracks the background fluid (radiation or matter) without dominating it. However, it remains unclear whether imposing such early-time non-phantom scaling dynamics is compatible with current observational data, particularly regarding the Hubble tension and late-time phantom-like preferences.

Methodology
The authors investigate a class of DDE models that admit non-phantom behavior at early times, specifically focusing on thawing, scaling–thawing, and effective fluid extensions. The study employs a minimally coupled scalar field with a double-exponential potential:
$$V(\phi) = V_1 e^{-\lambda_1 \phi/M_{Pl}} + V_2 e^{-\lambda_2 \phi/M_{Pl}}$$
where the steep term ($\lambda_2$) governs early-time scaling behavior, and the shallow term ($\lambda_1$) or an added cosmological constant/effective fluid governs late-time dynamics.

The authors analyze six distinct scenarios:

1. Thawing (TH): A single exponential potential with a shallow slope ($\lambda_1 < \sqrt{3}$).
2. Scaling+Thawing (SC+TH): The double-exponential potential described above.
3. Scaling+$\Lambda$ (SC+$\Lambda$): A steep exponential term combined with a cosmological constant.
4. Scaling+$w$CDM (SC+$w$): A steep exponential term combined with a constant equation of state (EoS) fluid.
5. Scaling+CPL (SC+CPL): A steep exponential term combined with a time-evolving CPL parametrization ($w(a) = w_0 + w_a(1-a)$).
6. $\Lambda$CDM: The standard baseline model.

Parameter estimation is performed using Markov Chain Monte Carlo (MCMC) techniques with the EMCEE sampler. The analysis utilizes a combined dataset comprising:

• Planck 2018 CMB distance priors (TT, TE, EE+lowE).
• DESI DR2 Baryon Acoustic Oscillation (BAO) measurements.
• Pantheon+ Type Ia Supernova (SNe Ia) sample.

Model performance is evaluated using the minimum chi-squared ($\chi^2_{min}$), reduced chi-squared, Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC) to assess statistical preference relative to $\Lambda$CDM while penalizing model complexity.

Key Contributions and Results

1. Stability of Background Parameters: Across all models, late-time background parameters ($\Omega_{m0}$, $h$, $\omega_b$, $r_d h$) remain stable and consistent with $\Lambda$CDM at the $1\sigma$ level. The inclusion of early-time scaling dynamics does not induce statistically significant shifts in these parameters.
2. Constraints on Potential Steepness ($\lambda_2$): Imposing non-phantom scaling dynamics at early times leads to strong lower bounds on the steepness of the exponential potential. The analysis finds $\lambda_2 \gtrsim 30$ for SC+TH, SC+$\Lambda$, and SC+$w$ models, and $\lambda_2 \gtrsim 20$ for the SC+CPL model.
3. Suppression of Early Dark Energy: These large values of $\lambda_2$ force the scalar field energy density to remain highly suppressed during radiation and matter domination. Specifically, the early dark energy fraction at matter–radiation equality is constrained to be below the percent level ($\Omega_\phi < 0.01$).
4. Hubble Tension: Due to the tight constraint on early dark energy density, the scaling behavior does not significantly alter the early expansion history or the sound horizon. Consequently, these models do not alleviate the $H_0$ tension.
5. Late-Time Phantom Preference: Time-dependent parametrizations, particularly CPL, show a $\sim 2\sigma$ preference for phantom evolution ($w < -1$) at low redshifts. However, this preference is sensitive to early-time assumptions.
6. Model Selection Penalties: While the SC+CPL model achieves the lowest $\chi^2_{min}$ (a $\Delta \chi^2 \approx -6$ improvement over $\Lambda$CDM), this improvement is insufficient to compensate for the increased model complexity. Both AIC and BIC penalize the inclusion of early-time scaling dynamics, rendering these models disfavored compared to $\Lambda$CDM.

Significance and Claims
The paper concludes that while late-time dynamical dark energy (specifically phantom-like evolution) may be mildly favored by current data, early-time non-phantom scaling behavior is strongly disfavoured. The requirement for a steep potential to maintain scaling dynamics effectively suppresses the early dark energy component to levels that cannot resolve the Hubble tension.

The authors assert that viable extensions of $\Lambda$CDM capable of accommodating both early and late-time dark energy dynamics likely require departures from straightforward scaling behavior or the introduction of physical mechanisms beyond the simple scalar field frameworks explored in this work. The study highlights a tension between the theoretical appeal of early scaling solutions and the observational constraints imposed by current cosmological data.