Study of the cosmological tensions and DESI-DR2 in the framework of the Little Rip model

This paper investigates the Little Rip dark energy model as a potential solution to the $H_0$ and $S_8$ cosmological tensions using recent observational data, including DESI-DR2, and finds that while the model reduces the Hubble tension to below $3\sigma$with early-universe data, Bayesian evidence indicates it only provides a statistically improved fit for CMB data compared to the standard$\Lambda$CDM model.

The Gist

Here is an explanation of the paper, translated into everyday language with some creative analogies.

The Big Picture: A Cosmic Tug-of-War

Imagine the universe as a giant, expanding balloon. For decades, scientists have been trying to figure out exactly how fast this balloon is inflating. They have two main ways of measuring it:

1. The "Baby Photo" Method (Early Universe): They look at the oldest light in the universe (the Cosmic Microwave Background, or CMB), which is like a baby photo of the cosmos. Based on this, they calculate the expansion rate to be about 67.
2. The "Adult Photo" Method (Late Universe): They look at exploding stars (Supernovae) and nearby galaxies right now. Based on this, they calculate the expansion rate to be about 73.

The Problem: These two numbers don't match. It's like measuring your height as a baby and getting 3 feet, but measuring yourself as an adult and getting 7 feet, even though you grew normally. In physics, this mismatch is called the Hubble Tension. There is also a second tension called the S8 Tension, which is about how "clumpy" the universe is (how much matter is bunched together vs. spread out).

The standard model of cosmology (called $\Lambda$CDM) is like a very reliable, old recipe book. It works great for most things, but it can't explain why these two measurements are so different.

The New Idea: The "Little Rip" Model

The authors of this paper decided to test a new recipe. They call it the Little Rip (LR) model.

• The Standard Recipe ($\Lambda$CDM): Imagine the universe's expansion is driven by a constant force, like a car cruising on cruise control. It expands steadily forever.
• The Little Rip Recipe: Imagine the expansion is driven by a force that gets slightly stronger over time, like a car that slowly presses the gas pedal harder and harder. It doesn't explode immediately (that would be the "Big Rip"), but it keeps accelerating until, eventually, the fabric of space-time gets torn apart.

This new recipe has one extra ingredient, a parameter called $\beta$ (beta). Think of $\beta$ as the "acceleration knob." If you turn it one way, the universe accelerates faster; turn it the other way, and it slows down.

The Experiment: Testing the Recipes

The scientists took the latest data from the most powerful telescopes and surveys available (including the new DESI data, which is like a high-definition map of millions of galaxies) and ran them through a computer simulation (MCMC). They asked: "Does the Little Rip model fit the data better than the old standard recipe?"

Here is what they found:

1. The "Baby Photo" vs. The "Adult Photo"

• When looking only at the "Baby Photo" (CMB data): The Little Rip model did a great job! It adjusted the numbers so that the tension between the early and late universe measurements dropped significantly. It was like the model found a way to make the baby photo and the adult photo look more consistent.
• When mixing in the "Adult Photo" (Supernova data): The model started to struggle. When they combined the old light with the new star data, the Little Rip model actually made the tension worse or just as bad as the standard model.

2. The "Clumpiness" Test (S8 Tension)

They also checked if the model could fix the "clumpiness" problem. They found that the Little Rip model could predict a level of clumpiness that matched some local surveys, but it still had trouble matching the "Baby Photo" data perfectly.

3. The Verdict: Which Recipe Wins?

To decide which model is better, the scientists used two statistical judges:

• The "Penalty" Judge (AIC): This judge says, "If you add a new ingredient (like $\beta$) to your recipe, you better make the dish taste much better, or I'm going to penalize you."
• The "Evidence" Judge (Bayes Factor): This judge looks at the whole picture, including how likely the new ingredient was to be chosen in the first place.

The Result:

• For the "Baby Photo" data alone: The Little Rip model won. The evidence was strong that this new recipe was better.
• For the "Mixed" data (Baby + Adult + Galaxy maps): The standard recipe ($\Lambda$CDM) won by a landslide. The Little Rip model didn't improve things enough to justify adding the extra "acceleration knob."

The Twist: The Universe is Changing Its Mind

One of the most interesting findings was about the direction of the "acceleration knob" ($\beta$).

• When they looked at the early universe, the knob pointed toward a "phantom" state (accelerating faster).
• But when they added the new DESI data (the latest galaxy maps), the knob flipped to the other side! It suggested the universe is actually behaving like Quintessence (a type of energy that is slightly less aggressive than a cosmological constant).

It's as if the universe is telling us, "I'm not just a simple constant, and I'm not just a runaway accelerator. I'm something in between, and I change my behavior depending on how you look at me."

The Conclusion

The paper concludes that while the Little Rip model is a fascinating idea that can solve the Hubble tension if you only look at the early universe, it fails to solve the problem when you look at the whole picture (combining early and late data).

In simple terms: The new model is a clever trick that works in a specific test, but when you throw all the real-world data at it, the old, reliable standard model ($\Lambda$CDM) still holds up the best. The universe remains a bit of a mystery, and we still need to figure out why our two best ways of measuring its speed don't agree.

Technical Summary

Here is a detailed technical summary of the paper "Study of the cosmological tensions and DESI-DR2 in the framework of the Little Rip model".

1. Problem Statement

The paper addresses two major "tensions" in modern cosmology that challenge the standard $\Lambda$CDM model:

• The $H_0$ Tension: The significant discrepancy (currently $\sim 5\sigma$) between the Hubble constant ($H_0$) measured from early-universe Cosmic Microwave Background (CMB) data (e.g., Planck, $H_0 \approx 67.4$ km s$^{-1}$ Mpc$^{-1}$) and late-universe distance ladder measurements (e.g., SH0ES, $H_0 \approx 73.0$ km s$^{-1}$ Mpc$^{-1}$).
• The $S_8$ Tension: The disagreement between the amplitude of matter fluctuations ($S_8$) derived from CMB data ($S_8 \approx 0.832$) and weak gravitational lensing surveys ($S_8 \approx 0.762$).

The authors investigate whether the Little Rip (LR) model, a specific dynamical dark energy scenario, can alleviate these tensions compared to the standard cosmological constant ($\Lambda$). The LR model is characterized by a future "abrupt event" where the universe expands indefinitely without a future singularity, driven by a dark energy equation of state (EoS) that crosses the phantom divide ($w < -1$).

2. Methodology

The study employs a rigorous statistical framework using the Markov Chain Monte Carlo (MCMC) method via the MontePython code interfaced with the Boltzmann solver CLASS.

• Theoretical Model:

  • The LR model is defined by the equation of state: $p_{de} = -(\rho_{de} + \beta\sqrt{\rho_{de}})$.
  • This introduces a single extra parameter, $\beta$, compared to $\Lambda$CDM.
  • Sign of $\beta$: $\beta > 0$ implies phantom dark energy ($w < -1$); $\beta < 0$ implies quintessence ($w > -1$).
  • The parameter space for $\beta$ is constrained to $[-0.05, 0.05]$ with a flat prior.

• Datasets Used:
  The analysis utilizes a comprehensive combination of the most recent observational data:

  • CMB: Planck 2018 temperature and polarization power spectra.
  • BAO: Data from 6dFGS, SDSS DR7, eBOSS DR16, and the latest DESI-DR2 (Data Release 2) measurements, including Bright Galaxy Samples, LRGs, ELGs, and Lyman-$\alpha$ forest data.
  • SNIa: PantheonPlus (1701 light curves), Union3 (2087 SNIa), and DES-Y5 (Dark Energy Survey Year 5).
  • Weak Lensing: KV450 and DES-Y1 combined constraints on $S_8$.

• Statistical Tools:

  • $\chi^2$ Analysis: To find best-fit parameters.
  • Akaike Information Criteria (AIC): To penalize models with extra parameters.
  • Bayesian Evidence (Bayes Factors): To compare model probabilities, accounting for prior volumes.

3. Key Contributions

• Integration of DESI-DR2: This is one of the first studies to test the Little Rip model against the latest DESI-DR2 BAO data combined with CMB and SNIa.
• Systematic Tension Analysis: The paper provides a comparative analysis of how the LR model affects $H_0$ and $S_8$ tensions across different dataset combinations (Early vs. Late universe data).
• Model Discrimination: It moves beyond simple $\chi^2$ minimization to use AIC and Bayes factors, offering a more robust statistical verdict on whether the extra parameter $\beta$ is justified by the data.

4. Key Results

A. Constraints on the Hubble Tension ($H_0$)

• CMB Only: The LR model yields a positive $\beta$ ($\approx 0.13 \times 10^{-3}$), indicating a phantom regime. It significantly raises the lower limit of $H_0$ ($> 72.86$ km s$^{-1}$ Mpc$^{-1}$), reducing the tension with SH0ES to $< 1\sigma$. In contrast, $\Lambda$CDM remains at $\sim 4.3\sigma$.
• CMB + BAO: The tension is reduced to $\sim 2.8\sigma$ for the LR model (compared to $>4\sigma$ for $\Lambda$CDM).
• CMB + BAO + SNIa (PantheonPlus): When late-universe supernova data is added, the value of $\beta$ becomes negative, shifting the model toward a quintessence regime ($w > -1$). Consequently, the preferred $H_0$ drops to $\approx 67.1$, and the tension reverts to $\sim 4.5\sigma$.
• DESI-DR2 Combinations: Combining DESI-DR2 with CMB and SNIa (Union3, PantheonPlus, or DES-Y5) consistently results in a negative $\beta$, favoring a quintessence-like dark energy.

B. Constraints on the $S_8$ Tension

• The LR model fits the weak lensing data (KV450 + DES-Y1) well, yielding $S_8 \approx 0.77$, which is consistent with local measurements but in tension with Planck18 ($S_8 \approx 0.832$) at $\sim 2.9\sigma$.
• When constrained by CMB alone, the model predicts $S_8 \approx 0.776$, showing consistency with weak lensing data at the $0.05\sigma$ level.

C. Model Selection (AIC and Bayes Factors)

• $\Delta \chi^2$: The LR model generally shows a slightly better fit (lower $\chi^2$) than $\Lambda$CDM for CMB-only and CMB+DESI-DR2+DES-Y5 datasets.
• AIC: The improvement in fit is marginal. For most datasets, the penalty for the extra parameter $\beta$ outweighs the gain in likelihood, favoring $\Lambda$CDM.
• Bayes Factors ($\ln B$):
  • CMB Only: Strong evidence ($\ln B = +6.78$) favors the LR model.
  • Combined Datasets (CMB+BAO, CMB+DESI, CMB+SNIa): Strong evidence ($\ln B < -5$) favors the $\Lambda$CDM model.
  • Interpretation: The data from BAO and SNIa are not fully consistent with the specific dynamics of the LR model when combined with CMB, effectively driving $\beta$ toward zero (reducing the model to $\Lambda$CDM).

D. Physical Evolution

• The study shows that the sign of $\beta$ is highly sensitive to the dataset combination.
• Positive $\beta$ (Phantom): Favored by high-redshift CMB data alone.
• Negative $\beta$ (Quintessence): Favored when low-redshift data (BAO, SNIa, DESI) are included.
• The functional posterior of the equation of state $w_{de}(z)$ shows that the LR model is indistinguishable from $\Lambda$CDM for most scales, with deviations only becoming significant for specific negative $\beta$ values at large scales ($\ell < 50$) in the CMB spectrum.

5. Significance and Conclusion

The paper concludes that while the Little Rip model offers a theoretical mechanism to potentially resolve the $H_0$ tension (specifically when using CMB data alone or CMB+BAO), it fails to provide a statistically significant improvement over the standard $\Lambda$CDM model when the full suite of modern cosmological data (including DESI-DR2 and SNIa) is considered.

• Key Finding: The inclusion of late-universe data (BAO and SNIa) forces the LR model into a quintessence regime ($\beta < 0$), which lowers the inferred $H_0$ and reintroduces the tension with local measurements.
• Statistical Verdict: According to Bayesian evidence, the $\Lambda$CDM model is strongly preferred over the LR model for all combined datasets. The extra parameter $\beta$ is not sufficiently supported by the data to justify the increased model complexity, except in the specific case of CMB-only analysis.
• Implication: The results suggest that the Little Rip scenario, in its current formulation, is not the primary solution to the cosmological tensions when confronted with the most comprehensive observational constraints available, particularly the new DESI-DR2 data.