Predicting High-magnification Events in Microlensed Quasars in the Era of LSST Using Recurrent Neural Networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14310%2F25%3A00144416" target="_blank" >RIV/00216224:14310/25:00144416 - isvavai.cz</a>

Výsledek na webu

<a href="https://iopscience.iop.org/article/10.3847/1538-4357/adaebb" target="_blank" >https://iopscience.iop.org/article/10.3847/1538-4357/adaebb</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.3847/1538-4357/adaebb" target="_blank" >10.3847/1538-4357/adaebb</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Predicting High-magnification Events in Microlensed Quasars in the Era of LSST Using Recurrent Neural Networks

Popis výsledku v původním jazyce

Upcoming wide-field surveys, such as the Rubin Observatory's Legacy Survey of Space and Time (LSST), will monitor thousands of strongly lensed quasars over a 10 yr period. Many of these monitored quasars will undergo high-magnification events (HMEs) through microlensing, as the accretion disk crosses a caustic-places of infinite magnification. Microlensing allows us to map the inner regions of the accretion disk as it crosses a caustic, even at large cosmological distances. The observational cadences of LSST are not ideal for probing the inner regions of the accretion disk, so there is a need to predict HMEs as early as possible, to trigger high-cadence multiband or spectroscopic follow-up observations. Here, we simulate a diverse and realistic sample of 10 yr quasar microlensing light curves to train a recurrent neural network to predict HMEs before they occur, by classifying the locations of the peaks at each time step. This is the first deep-learning approach for predicting HMEs. We give estimates of how well we expect to predict HME peaks during LSST and benchmark how our metrics change with different cadence strategies. With LSST-like observations, we can predict approximately 55% of HME peaks, corresponding to tens to hundreds per year and a false-positive rate of around 20% compared to the total number of HMEs. Our network can be continuously applied throughout the LSST survey, providing crucial alerts for optimizing follow-up resources.

Název v anglickém jazyce

Predicting High-magnification Events in Microlensed Quasars in the Era of LSST Using Recurrent Neural Networks

Popis výsledku anglicky

Upcoming wide-field surveys, such as the Rubin Observatory's Legacy Survey of Space and Time (LSST), will monitor thousands of strongly lensed quasars over a 10 yr period. Many of these monitored quasars will undergo high-magnification events (HMEs) through microlensing, as the accretion disk crosses a caustic-places of infinite magnification. Microlensing allows us to map the inner regions of the accretion disk as it crosses a caustic, even at large cosmological distances. The observational cadences of LSST are not ideal for probing the inner regions of the accretion disk, so there is a need to predict HMEs as early as possible, to trigger high-cadence multiband or spectroscopic follow-up observations. Here, we simulate a diverse and realistic sample of 10 yr quasar microlensing light curves to train a recurrent neural network to predict HMEs before they occur, by classifying the locations of the peaks at each time step. This is the first deep-learning approach for predicting HMEs. We give estimates of how well we expect to predict HME peaks during LSST and benchmark how our metrics change with different cadence strategies. With LSST-like observations, we can predict approximately 55% of HME peaks, corresponding to tens to hundreds per year and a false-positive rate of around 20% compared to the total number of HMEs. Our network can be continuously applied throughout the LSST survey, providing crucial alerts for optimizing follow-up resources.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10308 - Astronomy (including astrophysics,space science)

Projekt

<a href="/cs/project/GM24-10599M" target="_blank" >GM24-10599M: Hvězdy v galaktických jádrech: vzájemný vztah s masivními černými dírami</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Astrophysical Journal

ISSN

0004-637X

e-ISSN

1538-4357

Svazek periodika

981

Číslo periodika v rámci svazku

1

Stát vydavatele periodika

GB - Spojené království Velké Británie a Severního Irska

Počet stran výsledku

12

Strana od-do

1-12

Kód UT WoS článku

001432863400001

EID výsledku v databázi Scopus

2-s2.0-85218858574