WMAP Constraints on varying $α$ and the Promise of Reionization

C. J. A. P. Martins, A. Melchiorri, G. Rocha, Roberto Trotta, P. P. Avelino, P. T. P. Viana

Abstract

We present up-to-date constraints on the value of the fine-structure constant at the epoch of decoupling from the recent observations made by the Wilkinson Microwave Anisotropy Probe (WMAP) satellite. In the framework of models we considered, a positive (negative) variation of the value of $\alpha$ at decoupling with respect to the present-day value is now bounded to be smaller than 2% (6%) at 95% confidence level. We point out that the existence of an early reionization epoch as suggested by the above measurements will, when more accurate cosmic microwave background polarization data is available, lead to considerably tighter constraints. The so-called 'reionization bump', in particular, will be extremely useful for this purpose. We find that constraints on $\alpha$ can in principle only be as small as $0.1 %$ using CMB data alone—tighter constraints will require further (non-CMB) priors.

Introduction

In the standard model of cosmology and particle physics, fundamental constants like the fine-structure constant ($\alpha \approx 1/137$) are typically treated as immutable. However, modern theoretical frameworks, including string theory and other unification models, often posit the existence of extra dimensions that could lead to variations in these constants over cosmic time. Testing the constancy of $\alpha$ at the epoch of decoupling ($z \sim 1100$) provides a unique window into the physics of the early universe. This paper utilizes the first-year data from the Wilkinson Microwave Anisotropy Probe (WMAP) to establish new, robust constraints on $\alpha$ variation using the Cosmic Microwave Background (CMB) as a standalone laboratory.

Methods

The analysis employs two complementary approaches to probe the stability of $\alpha$. First, a likelihood estimation was performed by comparing WMAP’s temperature and cross-polarization data against a grid of flat cosmological models. These models were generated using a modified version of CMBFAST, accounting for how a change in $\alpha$ at decoupling would shift the recombination history and, consequently, the acoustic peaks of the CMB power spectrum.

Second, a Fisher Matrix Analysis (FMA) was used to forecast the sensitivity of future missions. This included simulations for the Planck satellite's High Frequency Instrument (HFI) and a theoretical "Cosmic Variance Limited" (CVL) experiment to determine the ultimate physical limits of CMB-based constraints.

Breaking Degeneracies with the Reionization Bump

A significant challenge in CMB analysis is the degeneracy between different cosmological parameters. For instance, both a variation in $\alpha$ and changes in the optical depth ($\tau$) can affect the power spectrum. However, this study highlights that the "reionization bump"—a signature in the polarization spectrum caused by the first stars—provides a critical second timestamp that helps break these degeneracies.

Comparison of the effects of reionization ($\tau$, left panels) and varying fine-structure constant ($\zeta = \alpha/\alpha_0$, right panels) on the CMB temperature (TT) and polarization (EE) power spectra. Note how $\alpha$ shifts peak positions while reionization creates a distinct bump at low $l$ in the EE spectrum.

As shown in the figure above, while $\alpha$ primarily shifts the angular scale and position of the acoustic peaks, reionization produces a unique enhancement at large angular scales (low multipoles $l$) in the EE polarization spectrum. By accurately measuring both the peak positions and the reionization bump, researchers can disentangle the effects of fundamental physics from the history of cosmic reionization.

Results and Future Limits

The WMAP first-year data provides the tightest standalone CMB constraints to date. At a 95% confidence level, the value of $\alpha$ at decoupling is constrained to be within 94% to 101% of its current value ($0.94 < \alpha_{dec}/\alpha_0 < 1.01$). This result is competitive with previous multi-probe analyses that combined CMB data with external datasets like Big Bang Nucleosynthesis or supernova observations.

Likelihood distribution for the variation of the fine-structure constant $\alpha/\alpha_0$ based on WMAP first-year data, showing a peak near the standard value.

Looking forward, the study forecasts that the Planck satellite will improve these constraints to within 0.34%. However, there is a theoretical "0.1% wall"; even an ideal CVL experiment cannot constrain $\alpha$ beyond approximately 0.1% using CMB data alone due to inherent cosmic variance and remaining parameter degeneracies. Surpassing this limit will require the integration of high-precision non-CMB priors, such as measurements of the Hubble constant.

Conclusion

This work demonstrates that the CMB is a powerful, independent probe of fundamental physics. While current WMAP data limits $\alpha$ variation to the few-percent level, the upcoming generation of high-sensitivity polarization experiments will leverage the reionization bump to reach sub-percent precision. Although these limits remain coarser than those from quasar absorption spectra ($10^{-5}$), the CMB offers a unique and pristine view of the universe's physical laws just 380,000 years after the Big Bang.

Summary updated

Abstract

We present up-to-date constraints on the value of the fine-structure constant at the epoch of decoupling from the recent observations made by the Wilkinson Microwave Anisotropy Probe (WMAP) satellite. In the framework of models we considered, a positive (negative) variation of the value of $α$ at decoupling with respect to the present-day value is now bounded to be smaller than 2% (6%) at 95% confidence level. We point out that the existence of an early reionization epoch as suggested by the above measurements will, when more accurate cosmic microwave background polarization data is available, lead to considerably tighter constraints. The so-called `reionization bump', in particular, will be extremely useful for this purpose. We find that constraints on $α$ can in principle only be as small as $0.1 %$ using CMB data alone--tighter constraints will require further (non-CMB) priors.
BibTeX
@inproceedings{martins2004wmap,
  author = {C. J. A. P. Martins and A. Melchiorri and G. Rocha and Roberto Trotta and P. P. Avelino and P. T. P. Viana},
  title = {{WMAP Constraints on varying $α$ and the Promise of Reionization}},
  booktitle = {Physics},
  year = {2004},
  doi = {10.1016/j.physletb.2003.11.080},
}
Figure