Date of Award
Summer 8-22-2021
Degree Type
Thesis
Degree Name
Master of Science (MS)
Department
Physics
First Advisor
Jesus Pando, PhD
Second Advisor
Anuj Sarma, PhD
Third Advisor
Bernhard Beck-Winchatz, PhD
Abstract
One of the goals of astrophysics is to obtain a full understanding how the Universe is organized on large scales and how structure evolved. In this thesis we develop a method of detecting structure on Mpc scales by measuring the one-dimensional power spectrum of the transmitted ux in the Lyman- forest. The method is based on the wavelet packet transform (WPT), which has several advantages over the Fourier transform. This includes reduced noise, resulting in less data manipulation and scrubbing in the early stages of analysis. Another advantage is localization of outliers in the data, which allows the general trend of the power spectrum to be revealed despite potentially problematic data. We apply the method to the set of 54,468 quasar spectra from the third collaboration of the Sloan Digital Sky Survey (SDSS-III) Baryonic Oscillation Spectroscopic Survey (BOSS) data release 9 (DR9) catalog. This is intended to be a proof of concept to determine if the wavelet packet power spectrum is a valid technique to extract the power spectrum in order to detect matter density uctuations. Results are in good agreement with previous studies that used conventional Fourier techniques. The power spectrum vs velocity space plots show increasing power at smaller scales for both our results and earlier studies by [21] and [6]. We conclude that the wavelet packet power spectrum is a tool for detecting structure from transmitted ux in the Lyman- forest. The advantages the wavelet packet power spectrum over the Fourier transform method are it requires less data manipulation and minimizes noise and propagation of errors and outliers in the data. As a next step we propose applying the tool to the larger more recent SDSS IV eBOSS dataset.
Recommended Citation
Pero, Jason, "WAVELET PACKET POWER SPECTRUM OF THE SDSS LYMAN-ALPHA FOREST: A TOOL FOR LARGE-SCALE STRUCTURE DETECTION" (2021). College of Science and Health Theses and Dissertations. 385.
https://via.library.depaul.edu/csh_etd/385
SLP Collection
no