[Paper Review] Scratches from the Past: Inflationary Archaeology through Features in the Power Spectrum of Primordial Fluctuations

Anže Slosar, Xingang Chen+5 more|arXiv (Cornell University)|Mar 23, 2019

Cosmology and Gravitation Theories84 references57 citations

TL;DR

A science white paper arguing that features in the primordial power spectrum are generic in inflationary models and should be pursued with future CMB and LSS surveys to access a wide range of scales and modes.

ABSTRACT

Inflation may provide unique insight into the physics at the highest available energy scales that cannot be replicated in any realistic terrestrial experiment. Features in the primordial power spectrum are generically predicted in a wide class of models of inflation and its alternatives, and are observationally one of the most overlooked channels for finding evidence for non-minimal inflationary models. Constraints from observations of the cosmic microwave background cover the widest range of feature frequencies, but the most sensitive constraints will come from future large-scale structure surveys that can measure the largest number of linear and quasi-linear modes.

Motivation & Objective

• Motivate searching for features in the primordial power spectrum as a probe of high-energy inflationary physics and possible alternatives to inflation.
• Argue that features are generic in many models and representative of fundamental physics beyond the minimal slow-roll scenario.
• Promote a coordinated program of next-generation CMB and large-scale structure experiments to maximize accessible scales and modes.

Proposed method

• Classify feature models by generation mechanism: resonant, sharp, localized, and primordial standard clocks.
• Describe how features imprint oscillations or localized deviations in the primordial power spectrum P(k) and relate them to observable signatures.
• Discuss reconstruction approaches for P(k) including parametric templates and non-parametric methods (e.g., penalized likelihood, Bayesian, splines, PCA).
• Outline forecast framework for feature detectability using CMB and LSS, including linear features as a basis (feature spectrometer) and the synergy between probes.

Experimental results

Research questions

• RQ1 What types of features in the primordial power spectrum are predicted by inflation and its alternatives?
• RQ2 How do different feature classes (linear, logarithmic, sharp, resonant, clock signals) map onto observable CMB and LSS signatures?
• RQ3 What is the potential gain in constraining features from future CMB polarization, galaxy surveys, 21 cm surveys, and spectral distortions?
• RQ4 Can a feature spectrometer framework unify constraints across diverse probes and models?
• RQ5 To what extent can features point to new physics or guide model-building in the early universe?

Key findings

• Features are a generic outcome in many inflationary and alternative models and could reveal high-energy physics of the primordial era.
• Current CMB constraints show no significant detections of features, with amplitudes limited to the percent level relative to the scalar amplitude.
• Future surveys—especially large-scale structure and 21 cm observations—promise substantial gains in constraining a wide range of feature frequencies and amplitudes.
• CMB remains most powerful for the largest feature frequencies, while LSS will improve constraints across a broad range of frequencies due to three-dimensional mode sampling.
• A feature spectrometer perspective helps visualize combined CMB and LSS sensitivity to linear features and their decomposition into general oscillatory templates.
• Observables such as non-Gaussianities and higher-point statistics can corroborate feature interpretations and connect to the underlying physics.