scholarly journals On the Quality of Deep Representations for Kepler Light Curves Using Variational Auto-Encoders

Signals ◽  
2021 ◽  
Vol 2 (4) ◽  
pp. 706-728
Author(s):  
Francisco Mena ◽  
Patricio Olivares ◽  
Margarita Bugueño ◽  
Gabriel Molina ◽  
Mauricio Araya
Keyword(s):  
Time Series ◽  
Light Curve ◽  
Latent Variables ◽  
Mean Squared Error ◽  
Pearson Correlation ◽  
Light Curves ◽  
Physical Parameters ◽  
Learning Approaches ◽  
Light Curve Analysis

Light curve analysis usually involves extracting manually designed features associated with physical parameters and visual inspection. The large amount of data collected nowadays in astronomy by different surveys represents a major challenge of characterizing these signals. Therefore, finding good informative representation for them is a key non-trivial task. Some studies have tried unsupervised machine learning approaches to generate this representation without much effectiveness. In this article, we show that variational auto-encoders can learn these representations by taking the difference between successive timestamps as an additional input. We present two versions of such auto-encoders: Variational Recurrent Auto-Encoder plus time (VRAEt) and re-Scaling Variational Recurrent Auto Encoder plus time (S-VRAEt). The objective is to achieve the most likely low-dimensional representation of the time series that matched latent variables and, in order to reconstruct it, should compactly contain the pattern information. In addition, the S-VRAEt embeds the re-scaling preprocessing of the time series into the model in order to use the Flux standard deviation in the learning of the light curves structure. To assess our approach, we used the largest transit light curve dataset obtained during the 4 years of the Kepler mission and compared to similar techniques in signal processing and light curves. The results show that the proposed methods obtain improvements in terms of the quality of the deep representation of phase-folded transit light curves with respect to their deterministic counterparts. Specifically, they present a good balance between the reconstruction task and the smoothness of the curve, validated with the root mean squared error, mean absolute error, and auto-correlation metrics. Furthermore, there was a good disentanglement in the representation, as validated by the Pearson correlation and mutual information metrics. Finally, a useful representation to distinguish categories was validated with the F1 score in the task of classifying exoplanets. Moreover, the S-VRAEt model increases all the advantages of VRAEt, achieving a classification performance quite close to its maximum model capacity and generating light curves that are visually comparable to a Mandel–Agol fit. Thus, the proposed methods present a new way of analyzing and characterizing light curves.

scholarly journals High-Level Magnetic Activity on a Low-Mass Close Binary: GSC 02038-0293

2012 ◽  
Vol 29 (2) ◽  
pp. 150-160 ◽  
Author(s):  
H. A. Dal ◽  
E. Sipahi ◽  
O. Özdarcan
Keyword(s):  
Light Curve ◽  
Active Component ◽  
Magnetic Activity ◽  
Curve Analysis ◽  
Light Curves ◽  
Close Binary ◽  
Physical Parameters ◽  
Light Curve Analysis ◽  
Low Mass ◽  
High Level

AbstractTaking into account results obtained from light-curve analysis and out-of-eclipse analyses, we discuss the nature of GSC 02038-00293 and also its magnetic activity behaviour.We obtained light curves of the system during observing seasons 2007, 2008 and 2011. We obtained its secondary minimum clearly in I-band observations in 2008 for the first time. Analysing this light curve, we found the physical parameters of the components. The light-curve analysis indicates that the possible mass ratio of the system is 0.35. We obtained the remaining V-band light curves, extracting the eclipses. We modelled these remaining curves using the SPOTMODEL program and found possible spot configurations of the magnetically active component for each observing season. The models demonstrated that there are two active longitudes for the active component. The models reveal that both active longitudes migrate in the direction of decreasing longitude. We also examined the light curves in out-of-eclipse phases with respect to minimum and maximum brightness, amplitude, etc. The amplitude of the curves during out-of-eclipse phases varies in a sinusoidal way with a period of ∼8.9 yr; the mean brightness of the system is dramatically decreasing. The phases of the deeper minimum during out-of-eclipse periods exhibit a migration toward decreasing phase.

scholarly journals MCMCI: A code to fully characterise an exoplanetary system

2020 ◽  
Vol 635 ◽  
pp. A6 ◽  
Author(s):  
A. Bonfanti ◽  
M. Gillon
Keyword(s):  
Time Series ◽  
Light Curve ◽  
Radial Velocity ◽  
Global Analysis ◽  
Major Axis ◽  
Theoretical Models ◽  
Integrated Analysis ◽  
Physical Parameters ◽  
Light Curve Analysis ◽  
Semi Major Axis

Context. Useful information can be retrieved by analysing the transit light curve of a planet-hosting star or induced radial velocity oscillations. However, inferring the physical parameters of the planet, such as mass, size, and semi-major axis, requires preliminary knowledge of some parameters of the host star, especially its mass or radius, which are generally inferred through theoretical evolutionary models. Aims. We seek to present and test a whole algorithm devoted to the complete characterisation of an exoplanetary system thanks to the global analysis of photometric or radial velocity time series combined with observational stellar parameters derived either from spectroscopy or photometry. Methods. We developed an integrated tool called MCMCI. This tool combines the Markov chain Monte Carlo (MCMC) approach of analysing photometric or radial velocity time series with a proper interpolation within stellar evolutionary isochrones and tracks, known as isochrone placement, to be performed at each chain step, to retrieve stellar theoretical parameters such as age, mass, and radius. Results. We tested the MCMCI on the HD 219134 multi-planetary system hosting two transiting rocky super Earths and on WASP-4, which hosts a bloated hot Jupiter. Even considering different input approaches, a final convergence was reached within the code, we found good agreement with the results already stated in the literature and we obtained more precise output parameters, especially concerning planetary masses. Conclusions. The MCMCI tool offers the opportunity to perform an integrated analysis of an exoplanetary system without splitting it into the preliminary stellar characterisation through theoretical models. Rather this approach favours a close interaction between light curve analysis and isochrones, so that the parameters recovered at each step of the MCMC enter as inputs for purposes of isochrone placement.

scholarly journals An extensive photometric study of the Blazhko RR Lyrae star RZ Lyr*

2012 ◽  
Vol 423 (2) ◽  
pp. 993-1005 ◽  
Author(s):  
J. Jurcsik ◽  
Á. Sódor ◽  
G. Hajdu ◽  
B. Szeidl ◽  
Á. Dózsa ◽  
...  
Keyword(s):  
Light Curve ◽  
Light Curves ◽  
Stellar Structure ◽  
Time Base ◽  
Photometric Method ◽  
Physical Parameters ◽  
Component Solution ◽  
Rr Lyrae ◽  
Curve Variation ◽  
Type Variable

Abstract The analysis of recent, extended multicolour CCD and archive photoelectric, photographic and visual observations has revealed several important properties of RZ Lyr, an RRab-type variable exhibiting large-amplitude Blazhko modulation. On the time base of ∼110 yr, a strict anticorrelation between the pulsation- and modulation-period changes is established. The light curve of RZ Lyr shows a remarkable bump on the descending branch in the small-amplitude phase of the modulation, similarly to the light curves of bump Cepheids. We speculate that the stellar structure temporally suits a 4:1 resonance between the periods of the fundamental and one of the higher order radial modes in this modulation phase. The light-curve variation of RZ Lyr can be correctly fitted with a two-modulation-component solution; the 121-d period of the main modulation is nearly but not exactly four times longer than the period of the secondary modulation component. Using the inverse photometric method, the variations in the pulsation-averaged values of the physical parameters in different phases of both modulation components are determined.

scholarly journals New Data on the Eclipsing Binary V1848 Ori and Improved Orbital and Light Curve Solutions

2020 ◽  
Vol 29 (1) ◽  
pp. 72-80 ◽  
Author(s):  
Fatemeh Davoudi ◽  
Atila Poro ◽  
Fahri Alicavus ◽  
Afshin Halavati ◽  
Saeed Doostmohammadi ◽  
...  
Keyword(s):  
Binary System ◽  
Light Curve ◽  
Mass Function ◽  
Light Curves ◽  
Complete Analysis ◽  
Physical Parameters ◽  
Mass Ratio ◽  
Third Body ◽  
Eclipsing Binary

AbstractNew observations of the eclipsing binary system V1848 Ori were carried out using the V filter resulting in a determination of new times of minima and new ephemeris were obtained. We presented the first complete analysis of the system’s orbital period behavior and analysis of O-C diagram done by the GA and MCMC approaches in OCFit code. The O-C diagram demonstrates a sinusoidal trend in the data; this trend suggests a cyclic change caused by the LITE effect with a period of 10.57 years and an amplitude of 7.182 minutes. It appears that there is a third body with mass function of f (m3) = 0.0058 M⊙ in this binary system. The light curves were analyzed using the Wilson-Devinney code to determine some geometrical and physical parameters of the system. These results show that V1848 Ori is a contact W UMa binary system with the mass ratio of q = 0.76 and a weak fillout factor of 5.8%. The O’Connell effect was not seen in the light curve and there is no need to add spot.

scholarly journals Hydrodynamic Models of Solar and Stellar Flares

1989 ◽  
Vol 104 (1) ◽  
pp. 289-298
Author(s):  
Giovanni Peres
Keyword(s):  
High Resolution ◽  
Light Curve ◽  
Solar Flares ◽  
Impulsive Phase ◽  
Light Curves ◽  
Physical Parameters ◽  
The Sun ◽  
Hydrodynamic Models ◽  
X Ray ◽  
Stellar Flares

AbstractThis paper discusses the hydrodynamic modeling of flaring plasma confined in magnetic loops and its objectives within the broader scope of flare physics. In particular, the Palermo-Harvard model is discussed along with its applications to the detailed fitting of X-ray light curves of solar flares and to the simulation of high-resolution Caxix spectra in the impulsive phase. These two approaches provide complementary constraints on the relevant features of solar flares. The extension to the stellar case, with the fitting of the light curve of an X-ray flare which occurred on Proxima Centauri, demonstrates the feasibility of using this kind of model for stars too. Although the stellar observations do not provide the wealth of details available for the Sun, and, therefore, constrain the model more loosely, there are strong motivations to pursue this line of research: the wider range of physical parameters in stellar flares and the possibility of studying further the solar-stellar connection.

scholarly journals Comparison of the power-2 limb-darkening law from the STAGGER-grid to Kepler light curves of transiting exoplanets

2018 ◽  
Vol 616 ◽  
pp. A39 ◽  
Author(s):  
P. F. L. Maxted
Keyword(s):  
Light Curve ◽  
Probability Distributions ◽  
Binary Star ◽  
Magnetic Activity ◽  
Light Curves ◽  
Practical Implementation ◽  
Light Curve Analysis ◽  
Stagger Grid ◽  
Least Squares Fit ◽  
Limb Darkening

Context. Inaccurate limb-darkening models can be a significant source of error in the analysis of the light curves for transiting exoplanet and eclipsing binary star systems, particularly for high-precision light curves at optical wavelengths. The power-2 limb-darkening law, Iλ(µ) = 1 − c(1−µα), has recently been proposed as a good compromise between complexity and precision in the treatment of limb-darkening. Aims. My aim is to develop a practical implementation of the power-2 limb-darkening law and to quantify the accuracy of this implementation. Methods. I have used synthetic spectra based on the 3D stellar atmosphere models from the STAGGER-grid to compute the limb-darkening for several passbands (UBVRI, CHEOPS, TESS, Kepler, etc.). The parameters of the power-2 limb-darkening laws are optimized using a least-squares fit to a simulated light curve computed directly from the tabulated Iλ(μ) values. I use the transformed parameters h1 = 1 − c(1 − 2−α) and h2 = c2−α to directly compare these optimized limb-darkening parameters to the limb darkening measured from Kepler light curves of 16 transiting exoplanet systems. Results. The posterior probability distributions (PPDs) of the transformed parameters h1 and h2 resulting from the light curve analysis are found to be much less strongly correlated than the PPDs for c and α. The agreement between the computed and observed values of (h1, h2) is generally very good but there are significant differences between the observed and computed values for Kepler-17, the only star in the sample that shows significant variability between the eclipses due to magnetic activity (star spots). Conclusions. The tabulation of h1 and h2 provided here can be used to accurately model the light curves of transiting exoplanets. I also provide estimates of the priors that should be applied to transformed parameters h1 and h2 based on my analysis of the Kepler light curves of 16 stars with transiting exoplanets.

scholarly journals PELICAN: deeP architecturE for the LIght Curve ANalysis

2019 ◽  
Vol 627 ◽  
pp. A21 ◽  
Author(s):  
Johanna Pasquet ◽  
Jérôme Pasquet ◽  
Marc Chaumont ◽  
Dominique Fouchez
Keyword(s):  
Light Curve ◽  
Sampling Rate ◽  
Real Data ◽  
Curve Analysis ◽  
Irregular Sampling ◽  
Light Curves ◽  
Light Curve Analysis ◽  
Deep Architecture ◽  
Deep Fields

We developed a deeP architecturE for the LIght Curve ANalysis (PELICAN) for the characterization and the classification of supernovae light curves. It takes light curves as input, without any additional features. PELICAN can deal with the sparsity and the irregular sampling of light curves. It is designed to remove the problem of non-representativeness between the training and test databases coming from the limitations of the spectroscopic follow-up. We applied our methodology on different supernovae light curve databases. First, we tested PELICAN on the Supernova Photometric Classification Challenge for which we obtained the best performance ever achieved with a non-representative training database, by reaching an accuracy of 0.811. Then we tested PELICAN on simulated light curves of the LSST Deep Fields for which PELICAN is able to detect 87.4% of supernovae Ia with a precision higher than 98%, by considering a non-representative training database of 2k light curves. PELICAN can be trained on light curves of LSST Deep Fields to classify light curves of the LSST main survey, which have a lower sampling rate and are more noisy. In this scenario, it reaches an accuracy of 96.5% with a training database of 2k light curves of the Deep Fields. This constitutes a pivotal result as type Ia supernovae candidates from the main survey might then be used to increase the statistics without additional spectroscopic follow-up. Finally we tested PELICAN on real data from the Sloan Digital Sky Survey. PELICAN reaches an accuracy of 86.8% with a training database composed of simulated data and a fraction of 10% of real data. The ability of PELICAN to deal with the different causes of non-representativeness between the training and test databases, and its robustness against survey properties and observational conditions, put it in the forefront of light curve classification tools for the LSST era.

scholarly journals Light Curve Modeling of Superluminous Supernovae

2013 ◽  
Vol 9 (S296) ◽  
pp. 86-89
Author(s):  
Takashi Moriya ◽  
Sergei I. Blinnikov ◽  
Nozomu Tominaga ◽  
Naoki Yoshida ◽  
Masaomi Tanaka ◽  
...  
Keyword(s):  
Kinetic Energy ◽  
Light Curve ◽  
Mass Loss ◽  
Numerical Simulations ◽  
Radiation Energy ◽  
Light Curves ◽  
Physical Parameters ◽  
Radiation Hydrodynamics ◽  
One Dimensional ◽  
Curve Modeling

AbstractOrigins of superluminous supernovae (SLSNe) discovered by recent SN surveys are still not known well. One idea to explain the huge luminosity is the collision of dense CSM and SN ejecta. If SN ejecta is surrounded by dense CSM, the kinetic energy of SN ejecta is efficiently converted to radiation energy, making them very bright. To see how well this idea works quantitatively, we performed numerical simulations of collisions of SN ejecta and dense CSM by using one-dimensional radiation hydrodynamics code STELLA and obtained light curves (LCs) resulting from the collision. First, we show the results of our LC modeling of SLSN 2006gy. We find that physical parameters of dense CSM estimated by using the idea of shock breakout in dense CSM (e.g., Chevalier & Irwin 2011, Moriya & Tominaga 2012) can explain the LC properties of SN 2006gy well. The dense CSM's radius is about 1016 cm and its mass about 15 M⊙. It should be ejected within a few decades before the explosion of the progenitor. We also discuss how LCs change with different CSM and SN ejecta properties and origins of the diversity of H-rich SLSNe. This can potentially be a probe to see diversities in mass-loss properties of the progenitors. Finally, we also discuss a possible signature of SN ejecta-CSM interaction which can be found in H-poor SLSN.

scholarly journals Synthetic Light Curve Analysis of the Close Binary Systems BX Andromedae and RR Leporis

1989 ◽  
Vol 107 ◽  
pp. 359-360
Author(s):  
R.G. Samec ◽  
R.E. Fuller ◽  
R. H. Kaitchuck ◽  
B. B. Bookmyer ◽  
D. R. Faulkner
Keyword(s):  
Light Curve ◽  
Binary Systems ◽  
Thermal Relaxation ◽  
Curve Analysis ◽  
Light Curves ◽  
Close Binary ◽  
Comparison Star ◽  
Short Time Scale ◽  
Light Curve Analysis ◽  
Short Period

AbstractUnpublished photoelectric observations of the systems BX And and RR Lep were subjected to light curve analysis using the Wilson-Devinney Code.The short-period eclipsing binary system BX And was observed on five nights in 1976 at the Morgan-Monroe station of the Goethe Link Observatory of Indiana University. The observations covering the eclipse portions of the light curves yielded four times of minimum light. A period study covering 89 years of observations confirms that a major period change took place about 1950. The system is suspected of being quite active on a short time scale. Standard magnitudes were derived for BX And and for the comparison star. The corrected color indices indicate that BX And and the comparison star BD+39° 476 are in the spectral range of F3-F5. The light curves, defined by 1092 observations in B, 1097 in V and 971 in the U filter are symmetric. The difference in the eclipse depths are quite large averaging 0.m46. A distinct de-reddening of the light curves occurs during the secondary eclipse. The first synthetic light curve solutions of the system were obtained. The solution of BX And indicates that the system consists of an F-type primary and a K-type secondary component in a state of shallow contact. This result is supported by the location of BX And on the Eggen period-color diagram for contact binaries. The large mass ratio, temperature disparity and period increase are found to be consistent with an early contact phase of thermal relaxation oscillations.

scholarly journals Revised Empirical Formulae for the Absolute Magnitudes and Intrinsic Colors of RRab Stars

2000 ◽  
Vol 176 ◽  
pp. 268-268 ◽  
Author(s):  
A. R. Walker ◽  
G. Kovács
Keyword(s):  
Light Curve ◽  
Absolute Magnitude ◽  
Large Datasets ◽  
Light Curves ◽  
Linear Functions ◽  
Physical Parameters ◽  
Fourier Amplitude ◽  
Color Indices ◽  
The Absolute ◽  
Absolute Magnitudes

AbstractThe relations between the physical parameters of the RRab stars and the Fourier parameters of their light curves have been reanalyzed on the basis of new CCD data. We show that the absolute magnitude MV and the dereddened color indices (B − V)0, (V − I)0 are linear functions of the period and of the A1 Fourier amplitude of the V light curve. Due to the large datasets, we are able to test the consistency of the formulae on independent sets. The present results are in good or fair agreement with our previous studies, despite the fewer number of significant parameters entering in the revised formulae.

Sign in / Sign up

Export Citation Format

Share Document