Bounds on Cosmic Strings from WMAP and SDSS

TL;DR

The paper constrains local cosmic strings using WMAP and SDSS by decomposing cosmological perturbations into adiabatic and string-sourced components and exploring a nine-parameter model that includes the string tension $G\\mu$ and wiggliness $\\alpha_r$ via Markov Chain Monte Carlo. It finds that cosmic strings can contribute a subdominant fraction of the CMB power, with an upper bound on the tension $G\\mu$ on the order of a few times $10^{-7}$ and a maximal string contribution of roughly a few percent to tens of percent depending on confidence level, all while keeping the overall cosmology close to $\\Lambda$CDM. The study also predicts B-mode polarization signatures from strings that could be detectable with future experiments and discusses how wiggliness and intercommutation affect the spectra. These results have implications for brane inflation scenarios and provide a framework for interpreting future CMB polarization observations in the context of cosmic-string models.

Abstract

We find the constraints from WMAP and SDSS data on the fraction of cosmological fluctuations sourced by local cosmic strings using a Markov Chain Monte Carlo (MCMC) analysis. In addition to varying the usual 6 cosmological parameters and the string tension ($μ$), we also varied the amount of small-scale structure on the strings. Our results indicate that cosmic strings can account for up to 7 (14)% of the total power of the microwave anisotropy at 68 (95)% confidence level. The corresponding bound on the string mass per unit length, within our string model, is $Gμ< 1.8 (2.7) \times 10^{-7}$ at 68 (95)% c.l., where this constraint has been altered from what appears below following the correction of errors in our cosmic string code outlined in a recent erratum, astro-ph/0604141. We also calculate the B-type polarization spectra sourced by cosmic strings and discuss the prospects of their detection.

Figures (7)

• Figure 1: The CMB TT and TE spectra (solid lines) sourced by cosmic strings with wiggliness parameter $\alpha_r=1.9$, as well as the adiabatic spectra for the same cosmological parameters (dashed lines) and WMAP's first year data. The string spectra are normalized so that the total TT power is the same for the two lines, which corresponds to $B=1$.
• Figure 2: The string-generated matter power spectrum (solid line) for the same parameters as in Fig. \ref{['stringcl']}, i.e., for $B=1$. The dashed line represents the linear spectrum from adiabatic perturbations at $z=0$.
• Figure 3: The one-dimensional projected PDFs for the 9 parameters varied by our Markov Chain Monte Carlo code; note that $\omega_B = \Omega_B h^2$, $\omega_M = \Omega_M h^2$. The solid line represents the PDFs for models where cosmic strings are included; the dashed line represents the PDFs for models with $B=0$, i.e. without cosmic strings. Each curve has been rescaled such that its area is unity. For each PDF the lightly shaded regions are excluded at the 68% confidence region; the dark regions are excluded at the 95% confidence region.
• Figure 4: The PDF for the Cosmic String weighting parameter, $B\equiv (G\mu / 2 \times 10^{-6})^2$.
• Figure 5: A contour plot of a separate set of MCMC results in the two-dimensional space defined by the parameters $B$ and $\alpha$. The contours are somewhat crude, but suggest that $\alpha$ is poorly constrained. At this level, there is no evidence for degeneracy between the parameters.