Local Primordial non-Gaussian Bias from Time Evolution

James M. Sullivan, Uroš Seljak · 2025

Read Local Primordial non-Gaussian Bias from Time Evolution by James M. Sullivan, Uroš Seljak (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-21736.txt.

Introduction

James Sullivan and Uroš Seljak tie a prized signal from the early cosmos to how galaxy counts change over time, in a 2025 paper. The signal, a slight skew in the newborn universe, would betray new physics at work back then. It leaves a mark in how galaxies cluster on vast scales, with a strength set by one factor. That factor turns out to track how the number of tracers like galaxies grows or fades with time.

For simple tracers whose counts hinge on peak height alone, the link is plain. For messier ones, a thought model of split-off mini-cosmoses gives the same tie, and the team checks it in simulations. They also show how to fix the factor when a survey's own picking rules warp the counts. Cleaner values of the factor should sharpen the hunt for the early signal in coming sky surveys.

Catalog record

CREATORJames M. Sullivan, Uroš Seljak
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE46 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-21736

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Local Primordial non-Gaussian Bias from Time Evolution James M. Sullivan, Uroš Seljak Abstract Primordial non-Gaussianity (PNG) is a signature of fundamental physics in the early universe that is probed by cosmological observations. It is well known that the local type of PNG generates a strong signal in the two-point function of large-scale structure tracers, such as galaxies. This signal, often termed “scale-dependent bias” is a generic feature of modulation of gravitational structure formation by a large-scale mode. It is less well-appreciated that the coefficient controlling this signal, b ϕ b_{\phi} , is closely connected to the time evolution of the tracer number density. This correspondence between time evolution and local PNG can be simply explained for a universal tracer whose mass function only depends on peak height, and more generally for non-universal tracers in the separate universe picture, which we validate in simulations. We also describe how to recover the bias of tracers subject to a survey selection function, and perform a simple demonstration on simulated galaxies. Since the local PNG amplitude in n − n- point statistics ( f NL f_{\rm NL} ) is largely degenerate with the coefficient b ϕ b_{\phi} , this proof of concept study demonstrates that galaxy survey data can allow for more optimal and robust extraction of local PNG information from upcoming surveys. Preprint: APS/123-QED I Introduction We are entering a golden age of large-scale structure (LSS) surveys. DESI 1 1 1 https://www.desi.lbl.gov/ and Euclid 2 2 2 https://www.cosmos.esa.int/web/euclid are currently operating, SPHEREx 3 3 3 https://spherex.caltech.edu/ has launched, LSST 4 4 4 https://www.lsst.org/ will begin taking data, and Spec-S5 5 5 5 https://www.spec-s5.org/ will spearhead future spectroscopic survey cosmology. These surveys span a range of galaxy number densities and volumes, but on the whole are tending toward increased sky coverage and higher redshifts. Primordial non-Gaussianity (PNG) is a prime target of these surveys, as any deviation from Gaussian initial conditions will be imprinted on tracers of LSS, such as galaxies, and will generally be easier to detect at the large scales these surveys will be sensitive to. While there are several well-understood ways in which the detailed physics of the early universe can generate non-Gaussian statistics, galaxy surveys are particularly well-suited to probe local PNG (with amplitude parameter f_{\rm NL}~=f_{\rm NL}^{(\rm loc)} ). A detection of the local PNG (LPNG) signal, under standard assumptions, would indicate the presence of multiple light fields during inflation. LSS surveys most sensitively detect LPNG through a large-scale signal in galaxy n- point statistics that cannot be generated by gravitational evolution [ 6 , 7 , 8 ] . This signal is often termed “scale-dependent bias” and can be understood as the coupling of modes in the initial conditions of structure formation (e.g., generating is a non-zero squeezed bispectrum, where one mode is much larger than the size of any gravitational displacement). LPNG bias has accumulated a long history of study [ 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 ] , but it was understood immediately that for universal tracers 6 6 6 We will frequently use the word “halo” or “galaxy”, since those are the cases of interest here, but all conclusions stated here generally apply to any LSS tracer. 7 7 7 With a universal mass function (UMF) depending only on peak height \nu(M,z)=\frac{\delta_{c}}{\sigma(M,z)} for critical overdensity \delta_{c} and amplitude of smoothed matter fluctuations \sigma the LPNG bias coefficient b_{\phi} is closely related to the the tracer Lagrangian bias b_{\phi}(b_{L})=2\delta_{c}b_{L}=2\delta_{c}(b_{E}-1) [ 9 , 10 ] . This relationship has been explored in dark-matter-only simulations and hydrodynamical simulations [ 20 , 21 , 22 , 23 , 24 ] , where it generally holds for dark matter halos selected by mass. A richer picture of b_{\phi} has emerged that extends the core connection to Lagrangian bias. From nearly the outset, possible deviations from the UMF relation have been captured by a free parameter p , the presence of which was motivated by the possibility of recent halo mergers [ 10 , 25 ] . But, recently, there has been an ambitious rally to physically model the deviation of b_{\phi} from its UMF relation (i.e., accounting for b_{\phi} “assembly bias”) [ 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 ] , as samples with large differences in b_{\phi} (with similar b_{1} , as typical of strong assembly bias splits) lead to improved constraints on f_{\rm NL}~ . However, even with the multi-tracer technique [ 35 , 36 ] and machine learning tools trained on simulations, previous efforts require a muscular assumption - that one knows b_{\phi} with high confidence. Relaxing this assumption can significantly degrade constraints on f_{\rm NL}~ [ 37 , 38 ] . The question we aim to answer in this paper is simple - can we wriggle out of strong assumptions about b_{\phi} by estimating b_{\phi} directly from the data? . We suggest that the answer to this question is in the affirmative, exploiting the close connection between b_{\phi} and the response of galaxy number density to the growth of structure, which can be related to the evolution bias b_{e} [ 39 , 40 , 41 , 42 ] . We first motivate the connection between time evolution and LPNG (Section II ), before validating this effect in N-body and hydrodynamical simulations (Section III ), showing that it generalizes beyond the universal case for galaxies in hydrodynamical simulations in addition to dark matter halos. We then address the important issue of selection effects (Section IV ), which must be properly accounted for before attempting to inform the value of b_{\phi} from tracer number density data. Finally we discuss possible extensions of the method develop here (Section V ) before concluding with an outlook towards future more optimal and robust f_{\rm NL}~ analyses (Section VI ). II Theory We present a simple derivation of the approximate equivalence of LPNG bias and evolution bias for universal tracers in Section II.1 and extend this to more general tracers in Section II.2 . II.1 Bias from universal tracers For a tracer whose abundance n_{t}(\nu) (in terms of comoving number density) is assumed to depend only on peak height \nu(M,z,\sigma_{8})\equiv\frac{\delta_{c}}{\sigma(M,z,\sigma_{8})} , it is simple to see the correspondence between Local PNG bias b_{\phi} and (time-) evolution bias b_{e} . Figure 1 gives a heuristic depiction of why this might be the case using field realizations 8 8 8 Here using an unrealistically large f_{NL}=5000 for illustration. . In a simple picture of tracer formation, halos/tracers form when the (smoothed) matter density field exceeds a certain collapse threshold. Local PNG (orange-dashed) acts to rescale the variance of the matter overdensity field via the large scale potential (green dotted), which affects tracer formation (some orange-dashed peaks cross the threshold when they did not in the Gaussian case [blue] and vice versa). One can roughly imagine that the orange-dashed curve is actually a future version of the blue curve in regions where the variance is enhanced - some peaks have grown while nearby “voids” have emptied out. We now show there is some merit to this story. Following Ref. [ 44 ] , for

Read the complete text