Publications
Table of Contents:
We present spectroscopic confirmation and lens modeling of the strong lensing system DESI-253.2534+26.8843, discovered in the Dark Energy Spectroscopic Instrument (DESI) Legacy Imaging Surveys data. This system consists of a massive elliptical galaxy surrounded by four blue images forming an Einstein Cross pattern. We obtained spectroscopic observations of this system using the Multi Unit Spectroscopic Explorer (MUSE) on ESO's Very Large Telescope (VLT) and confirmed its lensing nature. The main lens, which is the elliptical galaxy, has a redshift of zL1=0.636±0.001, while the spectra of the background source images are typical of a starburst galaxy and have a redshift of zs=2.597±0.001. Additionally, we identified a faint galaxy foreground of one of the lensed images, with a redshift of zL2=0.386. We employed the GIGA-Lens modeling code to characterize this system and determined the Einstein radius of the main lens to be θE=2.520′′+0.032−0.031, which corresponds to a velocity dispersion of σ = 379 ± 2 km/s. Our study contributes to a growing catalog of this rare kind of strong lensing systems and demonstrates the effectiveness of spectroscopic integral field unit observations and advanced modeling techniques in understanding the properties of these systems.
We develop a pipeline to perform a targeted lensed transient search. We apply this pipeline to 5807 strong lenses and candidates, identified in the literature, in the DESI Legacy Imaging Surveys Data Release 9 (DR9) footprint. For each system, we analyze every exposure in all observed bands (DECam g, r, and z). Our pipeline finds, groups, and ranks detections that are in sufficient proximity temporally and spatially. After the first round of inspection, for promising candidate systems, we further examine the newly available DR10 data (with additional i and Y bands). Here we present our targeted lensed supernova search pipeline and seven new lensed supernova candidates, including a very likely lensed supernova − probably a Type Ia − in a system with an Einstein radius of ~1.5′′.
We conduct a search for strongly lensed quasars in the Dark Energy Spectroscopic Instrument (DESI) Legacy Imaging Surveys (Dey et al. 2019) by applying an autocorrelation algorithm to ~ 5 million objects classified as quasars in the DESI Quasar Sample (Yeche et al. 2020). These systems are visually inspected and ranked. We present 436 new multiply-lensed and binary quasar candidates, 65 of which have redshifts from SDSS DR16. Redshifts are provided for an additional 17 candidates from the SuperNova Integral Field Spectrograph (SNIFS).
Storfer et al. 2022: New Strong Gravitational Lenses from the DESI Legacy Imaging Surveys Data Release 9
We have conducted a search for strong gravitational lensing systems in the Dark Energy Spectroscopic Instrument (DESI) Legacy Imaging Surveys Data Release 9. We use a deep residual neural network, trained on a compilation of known lensing systems. We have found 1895 lens candidates. Out of these, 1512 are identified for the first time. Combining the discoveries from this work, Huang et al. 2020, 2021 (335 and 1210 candidates respectively), the total number of strong lens candidates from the Legacy Surveys that we have discovered is 3057.
We present GIGA-Lens: a gradient-informed, GPU-accelerated Bayesian framework for modeling strong gravitational lensing systems, implemented in TensorFlow and JAX. The robustness, speed, and scalability offered by this framework make it possible to model the large number of strong lenses found in current surveys and present a very promising prospect for the modeling of O(10^5) lensing systems expected to be discovered in the era of the Vera C. Rubin Observatory, Euclid, and the Nancy Grace Roman Space Telescope.
We search in the Dark Energy Spectroscopic Instrument Legacy Imaging Surveys for new strong lensing systems by using deep residual neural networks, building on previous work presented in Huang et al. 2020. After applying our trained neural networks to the survey data, we visually inspect and rank images with probabilities above a threshold. Here we present 1210 new strong lens candidates.
We have performed a semi-automated search for strong gravitational lensing systems in the 9,000 deg2 Dark Energy Camera Legacy Surveys (DECaLS), part of the DESI Legacy Imaging Surveys. We adopted the deep residual neural network architecture developed by Lanusse et al.. We compiled a training sample that consists of observed non-lenses and known lensing systems. In this paper we present 335 candidate strong lensing systems, identified for the first time.
