Holismokes

Abstract

We present a systematic search for wide-separation (with Einstein radius θE ≳ 1.5″), galaxy-scale strong lenses in the 30 000 deg2 of the Pan-STARRS 3π survey on the Northern sky. With long time delays of a few days to weeks, these types of systems are particularly well-suited for catching strongly lensed supernovae with spatially-resolved multiple images and offer new insights on early-phase supernova spectroscopy and cosmography. We produced a set of realistic simulations by painting lensed COSMOS sources on Pan-STARRS image cutouts of lens luminous red galaxies (LRGs) with redshift and velocity dispersion known from the sloan digital sky survey (SDSS). First, we computed the photometry of mock lenses in gri bands and applied a simple catalog-level neural network to identify a sample of 1 050 207 galaxies with similar colors and magnitudes as the mocks. Second, we trained a convolutional neural network (CNN) on Pan-STARRS gri image cutouts to classify this sample and obtain sets of 105 760 and 12 382 lens candidates with scores of pCNN > 0.5 and > 0.9, respectively. Extensive tests showed that CNN performances rely heavily on the design of lens simulations and the choice of negative examples for training, but little on the network architecture. The CNN correctly classified 14 out of 16 test lenses, which are previously confirmed lens systems above the detection limit of Pan-STARRS. Finally, we visually inspected all galaxies with pCNN > 0.9 to assemble a final set of 330 high-quality newly-discovered lens candidates while recovering 23 published systems. For a subset, SDSS spectroscopy on the lens central regions proves that our method correctly identifies lens LRGs at z ∼ 0.1–0.7. Five spectra also show robust signatures of high-redshift background sources, and Pan-STARRS imaging confirms one of them as a quadruply-imaged red source at zs = 1.185, which is likely a recently quenched galaxy strongly lensed by a foreground LRG at zd = 0.3155. In the future, high-resolution imaging and spectroscopic follow-up will be required to validate Pan-STARRS lens candidates and derive strong lensing models. We also expect that the efficient and automated two-step classification method presented in this paper will be applicable to the ∼4 mag deeper gri stacks from the Rubin Observatory Legacy Survey of Space and Time (LSST) with minor adjustments.

Holismokes

“…The challenges associated with lensed SNe will be to find these systems amongst the millions of daily transient alerts from LSST and to analyse them quickly. Methods based on machine learning are being developed to overcome such challenges (e.g., Jacobs et al 2019;Avestruz et al 2019;Hezaveh et al 2017;Perreault Levasseur et al 2017;Pearson et al 2019;Cañameras et al 2020), and we are exploring these avenues in our forthcoming publications. lens and the source, respectively.…”

New High-quality Strong Lens Candidates with Deep Learning in the Kilo-Degree Survey

“…This can be done either by obtaining mock observations of fully simulated systems via ray-tracing of galaxies in hydrodynamical simulations (Lanusse et al 2018;He et al 2020) or by adding simulated arcs to images of real galaxies (e.g. Petrillo et al 2019a;Cañameras et al 2020). Here we follow this latter approach using KiDS images of galaxies to which we add simulated arcs, as already done in L+20.…”

Holismokes

“…Since these strong lens observations are very powerful, several large surveys including the Sloan Lens ACS (SLACS) survey (Bolton et al 2006;Shu et al 2017), the CFHTLS Strong Lensing Legacy Survey (SL2S; Cabanac et al 2007;Sonnenfeld et al 2015), the Sloan WFC Edge-on Late-type Lens Survey (SWELLS; Treu et al 2011), the BOSS Emission-Line Lens Survey (BELLS; Brownstein et al 2012;Shu et al 2016;Cornachione et al 2018), the Dark Energy Survey (DES; Dark Energy Survey Collaboration 2005; Tanoglidis et al 2021), the Survey of Gravitationally-lensed Objects in HSC Imaging (SuGOHI; Sonnenfeld et al 2018a;Wong et al 2018;Chan et al 2020;Jaelani et al 2020), and surveys in the Panoramic Survey Telescope and Rapid Response System (Pan-STARRS; e.g., Lemon et al 2018;Cañameras et al 2020) have been conducted to find lenses. So far we have detected several thousand lenses, but mainly from the lower redshift regime.…”

“…These methods are based on different identification properties, for instance on geometrical quantification (Bom et al 2017;Seidel & Bartelmann 2007), spectroscopic analysis (Baron & Poznanski 2017;Ostrovski et al 2017), or color cuts (Gavazzi et al 2014;Maturi et al 2014). Moreover, convolutional neural networks (CNNs) have also been extensively used in gravitational lens detection (e.g., Jacobs et al 2017;Petrillo et al 2017;Schaefer et al 2018;Lanusse et al 2018;Metcalf et al 2019;Cañameras et al 2020;Huang et al 2020) as they do not require any measurements of the lens properties. Once a CNN is trained, it can classify huge amounts of images in a very short time, and is thus very efficient.…”

“…We recently initiated the Highly Optimized Lensing Investigations of Supernovae, Microlensing Objects, and Kinematics of Ellipticals and Spirals (HOLISMOKES) program (Suyu et al 2020, hereafter HOLISMOKES I). After presenting our lens search project (Cañameras et al 2020, hereafter HOLISMOKES II), we present in this paper a CNN for modeling strong gravitationally lensed galaxies with ground-based imaging, taking advantage of four different filters and not applying lens light subtraction beforehand. In contrast to Pearson et al (2019), we use a mocked-up data set based on real observed galaxy cutouts since the performance of the CNN on real systems will be optimal when the mock systems used for training are as close to real lens observations as possible.…”