First Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Implications for Inflation

TL;DR

The paper uses the first-year WMAP data, in combination with small-scale CMB and LSS measurements, to confront inflationary predictions. It confirms a distinctive TE anti-correlation on large scales, signaling adiabatic superhorizon fluctuations and supporting an inflationary (accelerated-expansion) origin. By parameterizing scalar and tensor power spectra and applying slow-roll relations, the authors constrain key observables ($n_s$, $dn_s/d\ln k$, $r$, $A$) and test both single-field and double-field inflation scenarios, finding no strong tensor detection and a mild running preference. Among models, the minimally coupled $\lambda\phi^4$ potential is disfavored, while adiabatic initial conditions remain consistent; the data do not require isocurvature components, and the results provide meaningful constraints on the inflationary landscape and the smoothness of the inflaton potential. The work underscores the role of high-precision CMB measurements in narrowing viable inflationary theories and highlights the need for improved $\tau$ and high-$l$ measurements to break parameter degeneracies and sharpen model discrimination.

Abstract

We confront predictions of inflationary scenarios with the WMAP data, in combination with complementary small-scale CMB measurements and large-scale structure data. The WMAP detection of a large-angle anti-correlation in the temperature--polarization cross-power spectrum is the signature of adiabatic superhorizon fluctuations at the time of decoupling. The WMAP data are described by pure adiabatic fluctuations: we place an upper limit on a correlated CDM isocurvature component. Using WMAP constraints on the shape of the scalar power spectrum and the amplitude of gravity waves, we explore the parameter space of inflationary models that is consistent with the data. We place limits on inflationary models; for example, a minimally-coupled lambda phi^4 is disfavored at more than 3-sigma using WMAP data in combination with smaller scale CMB and large scale structure survey data. The limits on the primordial parameters using WMAP data alone are: n_s(k_0=0.002 Mpc^{-1})=1.20_{-0.11}^{+0.12}, dn/dlnk=-0.077^{+0.050}_{- 0.052}, A(k_0=0.002 Mpc}^{-1})=0.71^{+0.10}_{-0.11} (68% CL), and r(k_0=0.002 Mpc^{-1})<1.28 (95% CL).

Figures (9)

• Figure 1: Temperature-Polarization angular power spectrum. The large-angle TE power spectrum predicted in primordial adiabatic models (solid), primordial isocurvature models (dashed), and in causal scaling seed models (dotted). The WMAP TE data kogut/etal:2003 is shown for comparison, in bins of $\Delta l=10$.
• Figure 2: This figure shows $n_s$ as a function of $k$ for the WMAP (left), WMAPext+2dFGRS (middle) and WMAPext+2dFGRS$+$Lyman $\alpha$ (right) data sets. The mean (solid line) and the 68% (shaded area) and 95% (dashed lines) intervals are shown. The scales probed by WMAP, 2dFGRS and Lyman $\alpha\ $ are indicated on the figure.
• Figure 3: This set of figures shows part of the parameter space spanned by viable slow roll inflation models, with the WMAP 68% confidence region shown in dark blue and the 95% confidence region shown in light blue.
• Figure 4: This set of figures compares the fits from the WMAP (top row), WMAPext+2dFGRS (middle row) and WMAPext+2dFGRS+Ly$\alpha$ data (bottom row) to the predictions of specific classes of physically motivated inflation models. The color coding shows model classes referred to in the text: (A) red, (B) green, (C) magenta, (D) black. The dark and light blue regions are the joint 1--$\sigma$ and 2--$\sigma$ regions for the specified data sets (contrast this with the 1-d marginalized 1--$\sigma$ errors given in Table \ref{['table:single_field']}). We show only Monte Carlo models that are consistent with all three 2--$\sigma$ regions in each data set. This figure does not imply that the models not plotted are ruled out.
• Figure 5: This set of figures compares the fits from the WMAPext+2dFGRS+Ly$\alpha$ data to the predictions of all four classes of inflation models. The top row is Class A [red dots]. The second row is Class B [green dots]. The third row is Class C [magenta dots]. The bottom row is Class D [black dots]. The dark and light blue regions are the joint 1--$\sigma$ and 2--$\sigma$ regions for the WMAPext+2dFGRS+Ly$\alpha$ data. We show only Monte Carlo models that are consistent with 2--$\sigma$ regions in all panels. This figure does not imply that the models not plotted are ruled out.