Rates and Properties of Strongly Gravitationally Lensed Supernovae and their Host Galaxies in Time-Domain Imaging Surveys

TL;DR

This paper forecasts the discovery rates and properties of strongly gravitationally lensed supernovae in upcoming wide-field surveys like ZTF and LSST, highlighting their potential for astrophysics and cosmology.

Contribution

It introduces a comprehensive simulation framework to predict gLSNe rates and properties, including host galaxy modeling, for next-generation surveys.

Findings

ZTF and LSST will discover hundreds of gLSNe annually.

Most gLSN host galaxies will be multiply imaged, aiding lens modeling.

Simulated catalogs are publicly available for further research.

Abstract

Supernovae that are strongly gravitationally lensed (gLSNe) by galaxies are powerful probes of astrophysics and cosmology that will be discovered systematically by next-generation wide-field, high-cadence imaging surveys such as the Zwicky Transient Facility (ZTF) and the Large Synoptic Survey Telescope (LSST). Here we use pixel-level simulations that include dust, observing strategy, and multiple supernova subtypes to forecast the rates and properties of gLSNe that ZTF and LSST will find. Applying the resolution-insensitive discovery strategy of Goldstein et al. (2018), we forecast that ZTF (LSST) can discover 0.02 (0.79) 91bg-like, 0.17 (5.92) 91T-like, 1.22 (47.84) Type Ia, 2.76 (88.51) Type IIP, 0.31 (12.78) Type IIL, and 0.36 (15.43) Type Ib/c gLSNe per year. We also forecast that the surveys can discover at least 3.75 (209.32) Type IIn gLSNe per year, for a total of at least 8.60 (380.60) gLSNe per year under fiducial observing strategies. ZTF gLSNe have a median $z_s=0.9$, $z_l=0.35$, $\mu_\mathrm{tot}=30$, $\Delta t_\mathrm{max}= 10$ days, $\min(\theta)= 0.25^{\prime\prime}$, and $N_\mathrm{img} = 4$. LSST gLSNe are less compact and less magnified, with a median $z_s=1.0$, $z_l=0.4$, $\mu_\mathrm{tot}\approx6$, $\Delta t_\mathrm{max} = 25$ days, $\min(\theta)=0.6^{\prime\prime}$, and $N_\mathrm{img} = 2$. As the properties of lensed host galaxy arcs provide critical information for lens mass modeling, we develop a model of the supernova--host galaxy connection and use it to simulate realistic images of the supernova--host--lens systems. We find that the vast majority of gLSN host galaxies will be multiply imaged, enabling detailed constraints on lens models with sufficiently deep high-resolution imaging taken after the supernova has faded. We release the results of our simulations to the public as catalogs at this URL: http://portal.nersc.gov/project/astro250/glsne/.