Out of one, many: distinguishing time delays from lensed supernovae

Denissenya, Mikhail; Bag, Satadru; Kim, Alex; Linder, Eric V.; Shafieloo, Arman

doi:10.1093/mnras/stac143

Cited by 7 publications

(12 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The basic physical situation is of observation of only a single blended lightcurve from the combination of unresolved gravitationally lensed multiple images of a time varying source. We follow the notation of Bag et al (2021a); Denissenya et al (2022). The observed (blended) flux in a wavelength filter j is…”

Section: Deep Learning Approachmentioning

confidence: 99%

“…Unlensed systems have 1 image, while multiply lensed systems have 2 or 4 images (lensing gives an odd number of images but one lensed image is generally obscured by the lens galaxy or highly demagnified). Unlike Bag et al (2021a); Denissenya et al (2022), we do not input a form U (t), either an expansion about a given form or a free form bounded deviation from a given form. Instead we use a training set of generated, noisy, lensed and unlensed Type Ia supernova (SN Ia) lightcurves and use a convolutional neural net to classify the blended lightcurves as arising from an N images system.…”

Section: Deep Learning Approachmentioning

confidence: 99%

“…For training, we generate three data sets, each containing 10000 systems with successive time delays between images in the range [10,14] days, magnification ratios in [0.25, 4], and 5% measurement noise in flux (as in Bag et al (2021a); Denissenya et al (2022), for comparison), using the Hsiao supernova lightcurve template (Hsiao et al 2007) within sncosmo (Barbary et al 2020). Our LCsimulator code used to simulate unlensed and lensed systems is publicly available in a GitHub repository 1 .…”

Section: Training and Test Datamentioning

confidence: 99%

“…In all other sections we use a fiducial range of time delays ∆t = [10,14] for comparison to the previous work of Bag et al (2021a); Denissenya et al (2022). Recall that longer time delays would give more clearly resolved lightcurve peaks (hence images), while the previous techniques began to falter below ∆t = 10.…”

Section: Shorter Time Delaysmentioning

confidence: 99%

“…Such overlap poses three basic observational problems: 1) recognition of the lightcurve as a lensed supernova, when the lightcurve does not look like a standard supernova, 2) determination of the number of images contributing to the lightcurve, and 3) estimation of the time delays between all the images. Recent articles have addressed these by considering distortions of standard supernova lightcurves, by expanding in "crossing functions" (basically orthogonal polynomials) with arbitrary amplitudes (Bag et al 2021a), and by free form variation of lightcurves with constrained amplitudes (Denissenya et al 2022).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Deep Learning Unresolved Lensed Lightcurves

Denissenya,

Linder

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Gravitationally lensed sources may have unresolved or blended multiple images, and for time varying sources the lightcurves from individual images can overlap. We use convolutional neural nets to both classify the lightcurves as due to unlensed, double, or quad lensed sources and fit for the time delays. Focusing on lensed supernova systems with time delays ∆t 6 days, we achieve 100% precision and recall in identifying the number of images and then estimating the time delays to σ ∆t ≈ 1 day, with a 1000× speedup relative to our previous Monte Carlo technique. This also succeeds for flux noise levels ∼ 10%. For ∆t ∈ [2, 6] days we obtain 94-98% accuracy, depending on image configuration. We also explore using partial lightcurves where observations only start near maximum light, without the rise time data, and quantify the success.

show abstract

Section: Deep Learning Approachmentioning

confidence: 99%

Section: Deep Learning Approachmentioning

confidence: 99%

Section: Training and Test Datamentioning

confidence: 99%

Section: Shorter Time Delaysmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Deep Learning Unresolved Lensed Lightcurves

Denissenya,

Linder

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Searching for Strong Gravitational Lenses

Lemon,

Courbin,

et al. 2024

Space Sci Rev

View full text Add to dashboard Cite

Strong gravitational lenses provide unique laboratories for cosmological and astrophysical investigations, but they must first be discovered – a task that can be met with significant contamination by other astrophysical objects and asterisms. Here we review strong lens searches, covering various sources (quasars, galaxies, supernovae, FRBs, GRBs, and GWs), lenses (early- and late-type galaxies, groups, and clusters), datasets (imaging, spectra, and lightcurves), and wavelengths. We first present the physical characteristics of the lens and source populations, highlighting relevant details for constructing targeted searches. Search techniques are described based on the main lensing feature that is required for the technique to work, namely one of: (i) an associated magnification, (ii) multiple spatially-resolved images, (iii) multiple redshifts, or (iv) a non-zero time delay between images. To use the current lens samples for science, and for the design of future searches, we list several selection biases that exist due to these discovery techniques. We conclude by discussing the future of lens searches in upcoming surveys and the new population of lenses that will be discovered.

show abstract

Strongly Lensed Transient Sources: A Review

Liao¹,

Biesiada²,

Zhu³

2022

Chinese Phys. Lett.

View full text Add to dashboard Cite

The past decades have witnessed a lot of progress in gravitational lensing with two main targets: stars and galaxies (with active galactic nuclei). The success is partially attributed to the continuous luminescence of these sources making the detection and monitoring relatively easy. With the running of ongoing and upcoming large facilities/surveys in various electromagnetic and gravitational-wave bands, the era of time-domain surveys would guarantee constant detection of strongly lensed explosive transient events, for example, supernovae in all types, gamma ray bursts with afterglows in all bands, fast radio bursts and even gravitational waves. Lensed transients have many advantages over the traditional targets in studying the Universe and magnification effect helps to understand the transients themselves at high redshifts. In this review article, basing on the recent achievements in the literature, we summarize the methods of searching for different kinds of lensed transient signals, the latest results on detection and their applications in fundamental physics, astrophysics and cosmology. At the same time, we give supplementary comments as well as prospects of this emerging research direction that may help readers who are interested in entering this field.

show abstract

Out of one, many: distinguishing time delays from lensed supernovae

Cited by 7 publications

References 25 publications

Deep Learning Unresolved Lensed Lightcurves

Deep Learning Unresolved Lensed Lightcurves

Searching for Strong Gravitational Lenses

Strongly Lensed Transient Sources: A Review

Contact Info

Product

Resources

About