Aims. We inspect the four microlensing events KMT-2021-BLG-1968, KMT-2021-BLG-2010, KMT-2022-BLG-0371, and KMT-2022-BLG-1013, for which the light curves exhibit partially covered short-term central anomalies. We conduct detailed analyses of the events with the aim of revealing the nature of the anomalies. Methods. We tested various models that can explain the anomalies of the individual events, including the binary-lens (2L1S) and binary-source (1L2S) interpretations. Under the 2L1S interpretation, we thoroughly inspected the parameter space to determine the existence of degenerate solutions, and if they existed, we tested whether the degeneracy could be resolved. Results. We find that the anomalies in KMT-2021-BLG-2010 and KMT-2022-BLG-1013 are uniquely defined by planetary-lens interpretations with planet-to-host mass ratios of q ~ 2.8 × 10 −3 and ~1.6 × 10 −3 , respectively. For KMT-2022-BLG-0371, a planetary solution with a mass ratio q ~ 4 × 10 −4 is strongly favored over the other three degenerate 2L1S solutions with different mass ratios based on the χ 2 and relative proper motion arguments, and a 1L2S solution is clearly ruled out. For KMT-2021-BLG-1968, on the other hand, we find that the anomaly can be explained either by a planetary or a binary-source interpretation, making it difficult to firmly identify the nature of the anomaly. From the Bayesian analyses of the identified planetary events, we estimate that the masses of the planet and host are ( M p / M J , M h / M ⊙ ) = (1.07 −0.68 +1.15 , 0.37 −0.23 +0.40 ), (0.26 −0.11 +0.13 , 0.63 −0.28 +0.32 ), and (0.31 −0.16 +0.46 , 0.18 −0.10 +0.28 ) for KMT-2021-BLG-2010L, KMT-2022-BLG-0371L, and KMT-2022-BLG-1013L, respectively.
more »
« less
Dynamics signature based anomaly detection
Identifying anomalies, especially weak anomalies in constantly changing targets, is more difficult than in stable targets. In this article, we borrow the dynamics metrics and propose the concept of dynamics signature (DS) in multi-dimensional feature space to efficiently distinguish the abnormal event from the normal behaviors of a variable star. The corresponding dynamics criterion is proposed to check whether a star's current state is an anomaly. Based on the proposed concept of DS, we develop a highly optimized DS algorithm that can automatically detect anomalies from millions of stars' high cadence sky survey data in real-time. Microlensing, which is a typical anomaly in astronomical observation, is used to evaluate the proposed DS algorithm. Two datasets, parameterized sinusoidal dataset containing 262,440 light curves and real variable stars based dataset containing 462,996 light curves are used to evaluate the practical performance of the proposed DS algorithm. Experimental results show that our DS algorithm is highly accurate, sensitive to detecting weak microlensing events at very early stages, and fast enough to process 176,000 stars in less than 1 s on a commodity computer.
more »
« less
- Award ID(s):
- 2109988
- NSF-PAR ID:
- 10311659
- Date Published:
- Journal Name:
- Software: Practice and Experience
- ISSN:
- 0038-0644
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Aims. We conducted a project of reinvestigating the 2017–2019 microlensing data collected by high-cadence surveys with the aim of finding planets that were missed due to the deviations of planetary signals from the typical form of short-term anomalies. Methods. The project led us to find three planets, KMT-2017-BLG-2509Lb, OGLE-2017-BLG-1099Lb, and OGLE-2019-BLG-0299Lb. The lensing light curves of the events have a common characteristic: the planetary signals were produced by the crossings of faint source stars over the resonant caustics formed by giant planets located near the Einstein rings of host stars. Results. For all planetary events, the lensing solutions are uniquely determined without any degeneracy. It is estimated that the host masses are in the range of 0.45 ≲ M ∕ M ⊙ ≲ 0.59, which corresponds to early M to late K dwarfs, and thus the host stars are less massive than the Sun. On the other hand, the planets, with masses in the range of 2.1 ≲ M ∕ M J ≲ 6.2, are heavier than the heaviest planet of the Solar System, that is, Jupiter. The planets in all systems lie beyond the snow lines of the hosts, and thus the discovered planetary systems, together with many other microlensing planetary systems, support the idea that massive gas-giant planets are commonplace around low-mass stars. We discuss the role of late-time high-resolution imaging in clarifying resonant-image lenses with very faint sources.more » « less
-
Eclipsing post-common-envelope binaries are highly important for resolving the poorly understood, very short-lived common-envelope phase of stellar evolution. Most hot subdwarfs (sdO/Bs) are the bare helium-burning cores of red giants that have lost almost all of their hydrogen envelope. This mass loss is often triggered by common-envelope interactions with close stellar or even substellar companions. Cool companions to hot subdwarf stars such as late-type stars and brown dwarfs are detectable from characteristic light-curve variations – reflection effects and often eclipses. In the recently published catalog of eclipsing binaries in the Galactic Bulge and in the Asteroid Terrestrial-impact Last Alert System (ATLAS) survey, we discovered 125 new eclipsing systems showing a reflection effect seen by visual inspection of the light curves and using a machine-learning algorithm, in addition to the 36 systems previously discovered by the Optical Gravitational Lesing Experiment (OGLE) team. The Eclipsing Reflection Effect Binaries from Optical Surveys (EREBOS) project aims at analyzing all newly discovered eclipsing binaries of the HW Vir type (hot subdwarf + close, cool companion) based on a spectroscopic and photometric follow up to derive the mass distribution of the companions, constrain the fraction of substellar companions, and determine the minimum mass needed to strip off the red-giant envelope. To constrain the nature of the primary we derived the absolute magnitude and the reduced proper motion of all our targets with the help of the parallaxes and proper motions measured by the Gaia mission and compared those to the Gaia white-dwarf candidate catalog. It was possible to derive the nature of a subset of our targets, for which observed spectra are available, by measuring the atmospheric parameter of the primary, confirming that less than 10% of our systems are not sdO/Bs with cool companions but are white dwarfs or central stars of planetary nebula. This large sample of eclipsing hot subdwarfs with cool companions allowed us to derive a significant period distribution for hot subdwarfs with cool companions for the first time showing that the period distribution is much broader than previously thought and is ideally suited to finding the lowest-mass companions to hot subdwarf stars. The comparison with related binary populations shows that the period distribution of HW Vir systems is very similar to WD+dM systems and central stars of planetary nebula with cool companions. In the future, several new photometric surveys will be carried out, which will further increase the sample of this project, providing the potential to test many aspects of common-envelope theory and binary evolution.more » « less
-
more » « less
Abstract We used a convolutional neural network to infer stellar rotation periods from a set of synthetic light curves simulated with realistic spot-evolution patterns. We convolved these simulated light curves with real TESS light curves containing minimal intrinsic astrophysical variability to allow the network to learn TESS systematics and estimate rotation periods despite them. In addition to periods, we predict uncertainties via heteroskedastic regression to estimate the credibility of the period predictions. In the most credible half of the test data, we recover 10% accurate periods for 46% of the targets, and 20% accurate periods for 69% of the targets. Using our trained network, we successfully recover periods of real stars with literature rotation measurements, even past the 13.7 day limit generally encountered by TESS rotation searches using conventional period-finding techniques. Our method also demonstrates resistance to half-period aliases. We present the neural network and simulated training data, and introduce the software
butterpy used to synthesize the light curves using realistic starspot evolution. -
more » « less
ABSTRACT New time-domain surveys, such as the Vera C. Rubin Observatory Legacy Survey of Space and Time, will observe millions of transient alerts each night, making standard approaches of visually identifying new and interesting transients infeasible. We present two novel methods of automatically detecting anomalous transient light curves in real-time. Both methods are based on the simple idea that if the light curves from a known population of transients can be accurately modelled, any deviations from model predictions are likely anomalies. The first modelling approach is a probabilistic neural network built using Temporal Convolutional Networks (TCNs) and the second is an interpretable Bayesian parametric model of a transient. We demonstrate our methods’ ability to provide anomaly scores as a function of time on light curves from the Zwicky Transient Facility. We show that the flexibility of neural networks, the attribute that makes them such a powerful tool for many regression tasks, is what makes them less suitable for anomaly detection when compared with our parametric model. The parametric model is able to identify anomalies with respect to common supernova classes with high precision and recall scores, achieving area under the precision-recall curves above 0.79 for most rare classes such as kilonovae, tidal disruption events, intermediate luminosity transients, and pair-instability supernovae. Our ability to identify anomalies improves over the lifetime of the light curves. Our framework, used in conjunction with transient classifiers, will enable fast and prioritized followup of unusual transients from new large-scale surveys.
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1002/spe.3052
