Planck 2015 results. XVII. Constraints on primordial non-Gaussianity

TL;DR

Planck 2015 delivers stringent constraints on primordial non-Gaussianity by analyzing full-mission temperature and polarization maps with multiple estimators (KSW, modal, and binned) and robust simulations. The local, equilateral, and orthogonal f_NL limits from temperature and, preliminarily, from combined temperature and polarization data tighten the Gaussianity case and strengthen support for single-field slow-roll inflation, while constraining a broad class of alternative models including isocurvature, resonance, and excited initial states. Model-independent bispectrum reconstructions and trispectrum analyses (g_NL local) show no significant deviations from Gaussianity, corroborated by Minkowski functionals, and the results place meaningful bounds on a wide range of early-Universe scenarios. Polarization data, despite some residual systematics, consistently improve NG limits and provide complementary constraints with important implications for inflationary physics, though final polarization-only results remain preliminary. Overall, Planck 2015 findings reinforce ΛCDM with adiabatic Gaussian primordial perturbations and set the stage for even tighter tests with future CMB data.

Abstract

The Planck full mission cosmic microwave background(CMB) temperature and E-mode polarization maps are analysed to obtain constraints on primordial non-Gaussianity(NG). Using three classes of optimal bispectrum estimators - separable template-fitting (KSW), binned, and modal - we obtain consistent values for the local, equilateral, and orthogonal bispectrum amplitudes, quoting as our final result from temperature alone fNL^local=2.5+\-5.7, fNL^equil=-16+\-70 and fNL^ortho=-34+\-33(68%CL). Combining temperature and polarization data we obtain fNL^local=0.8+\-5.0, fNL^equil=-4+\-43 and fNL^ortho=-26+\-21 (68%CL). The results are based on cross-validation of these estimators on simulations, are stable across component separation techniques, pass an extensive suite of tests, and are consistent with Minkowski functionals based measurements. The effect of time-domain de-glitching systematics on the bispectrum is negligible. In spite of these test outcomes we conservatively label the results including polarization data as preliminary, owing to a known mismatch of the noise model in simulations and the data. Beyond fNL estimates, we present model-independent reconstructions of the CMB bispectrum and derive constraints on early universe scenarios that generate NG, including general single-field and axion inflation, initial state modifications, parity-violating tensor bispectra, and directionally dependent vector models. We also present a wide survey of scale-dependent oscillatory bispectra, and we look for isocurvature NG. Our constraint on the local primordial trispectrum amplitude is gNL^local=(-9.0+\-7.7)x10^4 (68%CL), and we perform an analysis of additional trispectrum shapes. The global picture is one of consistency with the premises of the LambdaCDM cosmology, namely that the structure we observe today was sourced by adiabatic, passive, Gaussian, and primordial seed perturbations.[abridged]

Figures (33)

• Figure 1: The skew-$C_\ell$ spectrum for the lensing-ISW effect (red line with data points), from the temperature map. The blue curve is the theoretically-expected spectrum. Note that the points beyond $\ell=1500$ are significantly correlated.
• Figure 2: The skew-$C_\ell$ spectrum for unclustered point sources (red line with data points), from the temperature map. The blue curve is the theoretical spectrum, given the amplitude determined with the KSW estimator.
• Figure 3: Map-by-map comparison of the results from the different estimators for local (left), equilateral (centre), and orthogonal (right) $f_\mathrm{NL}$ (taking the shapes to be independent), for the third set of simulations described in Sect. \ref{['Sec_valid_est_1']}, including both noise and a mask. Results are shown for T-only (top), E-only (centre), and the full combined T+E case (bottom). The legend for the estimators can be found in the top right figure. The horizontal solid line is the average value of all maps for KSW, and the dashed and dotted horizontal lines correspond to $\pm 1\sigma$ and $\pm 2\sigma$ deviations, respectively.
• Figure 4: Modal bispectrum reconstruction for Planck 2013 (top left) and 2015 (top right) temperature-only data, both using the SMICA maps. Here, we restrict the 2015 resolution to the same as 2013, using similar polynomials with $n_{\rm max} = 601$. The two bispectra are very close to being in complete agreement in the signal-dominated regime shown up to $\ell_{\rm max}= 1500$. In the lower panel, we show the Planck 2015temperature bispectrum at high resolution using the full $n_{\rm max} = 2001$ polynomial modes. Large-scale features in the top panels become subdivided but the main 2013 signals remain, notably a stronger measurement of the ISW-lensing signal (the regular oscillations in the squeezed limit).
• Figure 5: Modal reconstruction for the WMAP-9 bispectrum (left) and the Planck SMICA 2015 T-only bispectrum (right) plotted for the domain $\ell \le 450$, using identical isosurface levels. Here, we employed the full 2001 eigenmodes for both the Planck analysis at $\ell_\textrm{max} =2000$ and for WMAP-9 analysis at $\ell_\textrm{max} =600$, but for comparison purposes we have only used the first 600 eigenmodes in order to obtain a comparable resolution. The main features in the WMAP-9 bispectrum have counterparts in the Planck version, revealing an oscillatory pattern in the central region, as well as features on the tetrapyd surface. The WMAP-9 bispectrum has a much larger noise signal beyond $\ell =350$ than the more sensitive Planck experiment, leading to apparent residuals in this region.