Spectro-Imaging Forward Model of Red and Blue Galaxies

For the next generation of spectroscopic galaxy surveys, it is important to forecast their performances and to accurately interpret their large data sets. For this purpose, it is necessary to consistently simulate different populations of galaxies, in particular Emission Line Galaxies (ELGs), less u...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2020-06
Main Authors: Fagioli, Martina, Tortorelli, Luca, Herbel, Jörg, Zürcher, Dominik, Refregier, Alexandre, Amara, Adam
Format: Article
Language:English
Subjects:
Accuracy
Astronomical models
Computer simulation
Cosmology
Digital imaging
Galactic evolution
Photometry
Principal components analysis
Properties (attributes)
Sky surveys (astronomy)
Spectroscopy
Star & galaxy formation
Stars & galaxies
Statistical methods
Stellar age
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:For the next generation of spectroscopic galaxy surveys, it is important to forecast their performances and to accurately interpret their large data sets. For this purpose, it is necessary to consistently simulate different populations of galaxies, in particular Emission Line Galaxies (ELGs), less used in the past for cosmological purposes. In this work, we further the forward modeling approach presented in Fagioli et al. 2018, by extending the spectra simulator Uspec to model galaxies of different kinds with improved parameters from Tortorelli et al. 2020. Furthermore, we improve the modeling of the selection function by using the image simulator Ufig. We apply this to the Sloan Digital Sky Survey (SDSS), and simulate \(\sim157,000\) multi-band images. We pre-process and analyse them to apply cuts for target selection, and finally simulate SDSS/BOSS DR14 galaxy spectra. We compute photometric, astrometric and spectroscopic properties for red and blue, real and simulated galaxies, finding very good agreement. We compare the statistical properties of the samples by decomposing them with Principal Component Analysis (PCA). We find very good agreement for red galaxies and a good, but less pronounced one, for blue galaxies, as expected given the known difficulty of simulating those. Finally, we derive stellar population properties, mass-to-light ratios, ages and metallicities, for all samples, finding again very good agreement. This shows how this method can be used not only to forecast cosmology surveys, but it is also able to provide insights into studies of galaxy formation and evolution.
ISSN:2331-8422
DOI:10.48550/arxiv.2002.04039