Testing Inflation with Large Scale Structure: Connecting Hopes with Reality

Axionlike Origin of the Primordial Density Perturbation

“…This is in contrast to singlefield inflation which yields a much smaller local f NL , and also to a vanilla curvaton which produces f NL = −5/4 in the dominating limit (R → ∞). These values of non-Gaussianity are within reach of upcoming large-scale structure surveys [12]. We also remark that the axion lies within π/2 < |θ| < π and sources the red spectral tilt when the cosmological scales exit the horizon, and then rolls down typically to the region |θ| < π/2 by the end of inflation.…”

Low-redshift 21cm Cosmology in Canada

“…Thus, observations in the former regime (where future 21 cm surveys will have unparalleled performance) can be more easily related to the physics of the early universe. Constraints on two of the key observational targets for learning about this era-signatures in multi-point density correlation functions that depend on the field content of the early universe (so-called "primordial non-Gaussianity" Alvarez et al 2014), and sharp or oscillatory features in the primordial power spectrum that generically arise in many models of cosmic inflation (Chluba et al, 2015)-scale with the number of accessible linear Fourier modes (see Figure 1). 21 cm surveys, as the most cost-effective and scalable way of surveying large-scale structure, are ideally positioned to capitalize on this scaling and vastly improve our knowledge of the early universe, as verified in recent forecasts (Ansari et al, 2018).…”

Predictions for local PNG bias in the galaxy power spectrum and bispectrum and the consequences for $f_{\rm NL}$ constraints

“…In this way, the contribution scales as ∝ (b 1 − 1) f nl /k 2 , and since b 1 can in principle be constrained using the smaller-scale part of the power spectrum (where f nl contributes weakly), it then becomes possible to constrain f nl . The universality relation is adopted by almost all existing galaxy data constraints on f nl [16,[18][19][20][21][22][23][24][25] (the current tightest bound is f nl = −12 ± 21 (1σ) [25]), as well as in forecast studies [7,[26][27][28][29][30][31][32][33][34][35][36] for next-generation surveys. Despite its widespread adoption, there is however no reason to expect the universality relation to hold for real-life galaxy samples, and in fact, studies using N -body simulations have been indicating this to be the case already.…”