SearcharxivSearch

arXiv subjects

Joseph Tomlinson

Publications and source records attributed to Joseph Tomlinson.

3 recordsLinked to original sources

Spherical Bispectrum: A Novel Visualization Scheme For Facilitating Comparisons

Recent developments of Perturbation Theory (PT), specifically the Effective Field Theory of Large Scale Structure (EFTofLSS) and its equivalents, have proven powerful in analyzing galaxy clustering statistics such as the galaxy power spectrum and bispectrum. To further this pursuit, we have devised a novel spherical-bispectrum visualization scheme that collapses configuration dependencies to highlight the scale dependence of the bispectrum. The resulting one-dimensional curves facilitate the comparison between different bispectra, for example, from simulation and PT calculation. Using the new scheme, we present a quantitative analysis of the accuracy of PT modeling by comparing PT's analytical prediction to the result from a suite of Quijote simulations. Specifically, we determine $k_{\rm NL}$, the wavenunmber below which the analytical prediction matches well with the N-body result by inspecting both leading order (LO) and next-to-leading order (NLO) power spectrum and bispectrum at redshifts $z=0$, $0.5$, $1$, $2$, $3$. We also quantify the binning effect in Fourier space and show that an appropriate correction must be applied to the analytic predictions in order to compare them with the discrete Fourier transform results obtained from N-body-simulation or real data.

astro-ph.CO

Fast Calculation of Nonlinear Redshift-space Galaxy Power Spectrum Including Selection Bias

We present a fast implementation of the next-to-leading order (1-loop) redshift-space galaxy power spectrum by using FFTlog-based methods. [V. Desjacques, D. Jeong, and F. Schmidt, JCAP 1812 (12), 035] have shown that the 1-loop galaxy power spectrum in redshift space can be computed with 28 independent loop integrals with 22 bias parameters. Analytical calculation of the angular part of the loop integrals leaves the radial part in the form of a spherical Bessel transformation that is ready to be integrated numerically by using the FFTLog transformation. We find that the original 28 loop integrals can be solved with a total of 85 unique FFTLog transformations, yet leading to a few orders of magnitude speed up over traditional multi-dimensional integration. The code used in this work is publicly available at https://github.com/JosephTomlinson/GeneralBiasPk

astro-ph.CO

Efficient parallel algorithm for estimating higher-order polyspectra

Nonlinearities in the gravitational evolution, galaxy bias, and redshift-space distortion drive the observed galaxy density fields away from the initial near-Gaussian states. Exploiting such a non-Gaussian galaxy density field requires measuring higher-order correlation functions, or, its Fourier counterpart, polyspectra. Here, we present an efficient parallel algorithm for estimating higher-order polyspectra. Based upon the Scoccimarro estimator, the estimator avoids direct sampling of polygons by using the Fast-Fourier Transform (FFT), and the parallelization overcomes the large memory requirement of the original estimator. In particular, we design the memory layout to minimize the inter-CPU communications, which excels in the code performance.

astro-ph.CO