Search

The advent of large astronomical surveys has made available large and complex data sets. However, the process of discovery and interpretation of each potentially new astronomical source is, many times, still handcrafted. In this context, machine learning algorithms have emerged as a powerful tool to mine large data sets and lower the burden on the domain expert. Active learning strategies are specially good in this task. In this report, we used the PineForest algorithm to search for superluminous supernova (SLSN) candidates in the Zwicky Transient Facility. We showcase how the use of previously confirmed sources can provide important information to boost the convergence of the active learning algorithm. Starting from a data set of $\sim$14 million objects, and using 8 previously confirmed SLSN light curves as priors, we scrutinized 120 candidates and found 8 SLSN candidates, 2 of which have not been reported before (AT 2018moa and AT 2018mob). These results demonstrate how existing spectroscopic samples can be used to improve the efficiency of active learning strategies in searching for rare astronomical sources., Comment: 16 pages, 2 tables, 8 figures

We present coniferest, an open source generic purpose active anomaly detection framework written in Python. The package design and implemented algorithms are described. Currently, static outlier detection analysis is supported via the Isolation forest algorithm. Moreover, Active Anomaly Discovery (AAD) and Pineforest algorithms are available to tackle active anomaly detection problems. The algorithms and package performance are evaluated on a series of synthetic datasets. We also describe a few success cases which resulted from applying the package to real astronomical data in active anomaly detection tasks within the SNAD project., Comment: 13 pages, 1 figure

In the task of anomaly detection in modern time-domain photometric surveys, the primary goal is to identify astrophysically interesting, rare, and unusual objects among a large volume of data. Unfortunately, artifacts -- such as plane or satellite tracks, bad columns on CCDs, and ghosts -- often constitute significant contaminants in results from anomaly detection analysis. In such contexts, the Active Anomaly Discovery (AAD) algorithm allows tailoring the output of anomaly detection pipelines according to what the expert judges to be scientifically interesting. We demonstrate how the introduction real-bogus scores, obtained from a machine learning classifier, improves the results from AAD. Using labeled data from the SNAD ZTF knowledge database, we train four real-bogus classifiers: XGBoost, CatBoost, Random Forest, and Extremely Randomized Trees. All the models perform real-bogus classification with similar effectiveness, achieving ROC-AUC scores ranging from 0.93 to 0.95. Consequently, we select the Random Forest model as the main model due to its simplicity and interpretability. The Random Forest classifier is applied to 67 million light curves from ZTF DR17. The output real-bogus score is used as an additional feature for two anomaly detection algorithms: static Isolation Forest and AAD. While results from Isolation Forest remained unchanged, the number of artifacts detected by the active approach decreases significantly with the inclusion of the real-bogus score, from 27 to 3 out of 100. We conclude that incorporating the real-bogus classifier result as an additional feature in the active anomaly detection pipeline significantly reduces the number of artifacts in the outputs, thereby increasing the incidence of astrophysically interesting objects presented to human experts., Comment: 8 pages, 6 figures

We study the thermal stability of non-self-gravitating turbulent $\alpha$ discs around supermassive black holes (SMBHs) to test a new type of high-amplitude active galactic nuclei (AGN) flares. On calculating discs structures, we compute the critical points of stability curves for discs around SMBH, which cover a wide range of accretion rates and resemble the shape of a $\xi$ curve. We find that there are values of the disc parameters that favour the transition of a disc ring from a recombined cool state to a hot, fully ionised, advection dominated, geometrically thick state with higher viscosity parameter $\alpha$. For SMBH with masses $\sim 10^6-10^8 M_\odot$, such a flare can occur in the geometrically thin and optically thick neutral disc with convective energy transfer through the disc thickness surrounding a radiatively inefficient accretion flow. When self-gravity effects are negligible, the duration of a flare and the associated mass exhibit a positive correlation with the truncation radius of the geometrically thin disc prior to the flare. According to our rough estimates, $\sim 4-3000 M_\odot$ can be involved in a giant flare, i.e. can be accreted or entrained with an outflow lasting 1 to 400 years, if the flare is triggered somewhere between $60$ and $600$ gravitational radii in a disc around SMBH with $10^7 M_\odot$. The accretion rate on SMBH peaks at a super-Eddington value about ten times faster. The peak effective disc temperature at the trigger radius is $\sim 10^5\,$K, but it can be obscured by an optically thick outflow that reprocesses the emission to longer wavelengths. Such a transfer of disc state could trigger a massive outburst, similar to that following a tidal disruption event., Comment: 10 pages with 10 figures, 1 appendix with 4 figures; submitted to A&A; Comments are welcome

Most of the stars in the Universe are M spectral class dwarfs, which are known to be the source of bright and frequent stellar flares. In this paper, we propose new approaches to discover M-dwarf flares in ground-based photometric surveys. We employ two approaches: a modification of a traditional method of parametric fit search and a machine learning algorithm based on active anomaly detection. The algorithms are applied to Zwicky Transient Facility (ZTF) data release 8, which includes the data from the ZTF high-cadence survey, allowing us to reveal flares lasting from minutes to hours. We analyze over 35 million ZTF light curves and visually scrutinize 1168 candidates suggested by the algorithms to filter out artifacts, occultations of a star by an asteroid, and other types of known variable objects. The result of this analysis is the largest catalogue of ZTF flaring stars to date, representing 134 flares with amplitudes ranging from -0.2 to -4.6 magnitudes, including repeated flares. Using Pan-STARRS DR2 colors, we assign a spectral subclass to each object in the sample. For 13 flares with well-sampled light curves and available geometric distances from Gaia DR3, we estimate the bolometric energy. This research shows that the proposed methods combined with the ZTF's cadence strategy are suitable for identifying M-dwarf flares and other fast transients, allowing for the extraction of significant astrophysical information from their light curves., Comment: 19 pages, 6 figures, 1 table, 1 appendix

Theoretical models of accretion discs and observational data indicate that the X-ray emission from the inner parts of an accretion disc can irradiate its outer regions and induce a thermal wind, which carries away the mass and angular momentum from the disc. Our aim is to investigate the influence of the thermal wind on the outburst light curves of black hole X-ray binary systems. We carry out numerical simulations of a non-stationary disc accretion with wind using upgraded open code freddi. We assume that the wind launches only from the ionised part of the disc and may turn off if the latter shrinks fast enough. Our estimates of the viscosity parameter $\alpha$ are shifted downward compared to a scenario without a wind. Generally, correction of $\alpha$ depends on the spectral hardness of central X-rays and the disc outer radius, but unlikely to exceed a factor of 10 in the case of a black hole low-mass X-ray binary (BH LMXB). We fit 2002 outburst of BH LMXB 4U 1543-47 taking into account the thermal wind. The mass loss in the thermal wind is of order of the accretion rate on the central object at the peak of the outburst. New estimate of the viscosity parameter $\alpha$ for the accretion disc in this system is about two times lower than the previous one. Additionally, we calculate evolution of the number of hydrogen atoms towards 4U 1543-47 due to the thermal wind from the hot disc., Comment: 19 pages, 22 figures, accepted for publication in MNRAS

We present Rainbow, a physically motivated framework which enables simultaneous multi-band light curve fitting. It allows the user to construct a 2-dimensional continuous surface across wavelength and time, even in situations where the number of observations in each filter is significantly limited. Assuming the electromagnetic radiation emission from the transient can be approximated by a black-body, we combined an expected temperature evolution and a parametric function describing its bolometric light curve. These three ingredients allow the information available in one passband to guide the reconstruction in the others, thus enabling a proper use of multi-survey data. We demonstrate the effectiveness of our method by applying it to simulated data from the Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC) as well as real data from the Young Supernova Experiment (YSE DR1). We evaluate the quality of the estimated light curves according to three different tests: goodness of fit, time of peak prediction and ability to transfer information to machine learning (ML) based classifiers. Results confirm that Rainbow leads to equivalent (SNII) or up to 75% better (SN Ibc) goodness of fit when compared to the Monochromatic approach. Similarly, accuracy when using Rainbow best-fit values as a parameter space in multi-class ML classification improves for all classes in our sample. An efficient implementation of Rainbow has been publicly released as part of the light curve package at https://github.com/light-curve/light-curve-python. Our approach enables straight forward light curve estimation for objects with observations in multiple filters and from multiple experiments. It is particularly well suited for situations where light curve sampling is sparse., Comment: 14 pages, 15 figures, submitted to A&A

Radial structure of accretion discs around compact objects is often described using analytic approximations which are derived from averaging or integrating vertical structure equations. For non-solar chemical composition, partial ionization, or for supermassive black holes, this approach is not accurate. Additionally, radial extension of `analytically-described' disc zones is not evident in many cases. We calculate vertical structure of accretion discs around compact objects, with and without external irradiation, with radiative and convective energy transport taken into account. For this, we introduce a new open Python code, allowing different equations of state (EoS) and opacity laws, including tabular values. As a result, radial structure and stability `S-curves' are calculated for specific disc parameters and chemical composition. In particular, based on more accurate power-law approximations for opacity in the disc, we supply new analytic formulas for the farthest regions of the hot disc around stellar-mass object. On calculating vertical structure of a self-irradiated disc, we calculate a self-consistent value of the irradiation parameter $C_{\rm irr}$ for stationary $\alpha$-disc. We find that, for a fixed shape of the X-ray spectrum, $C_{\rm irr}$ depends weakly on the accretion rate but changes with radius, and the dependence is driven by the conditions in the photosphere and disc opening angle. The hot zone extent depends on the ratio between irradiating and intrinsic flux: corresponding relation for $T_{\rm irr,\, crit}$ is obtained., Comment: Accepted for publication in MNRAS, 15 pages, 15 figures, 3 appendix

We report the automatic detection of 11 transients (7 possible supernovae and 4 active galactic nuclei candidates) within the Zwicky Transient Facility fourth data release (ZTF DR4), all of them observed in 2018 and absent from public catalogs. Among these, three were not part of the ZTF alert stream. Our transient mining strategy employs 41 physically motivated features extracted from both real light curves and four simulated light curve models (SN Ia, SN II, TDE, SLSN-I). These features are input to a k-D tree algorithm, from which we calculate the 15 nearest neighbors. After pre-processing and selection cuts, our dataset contained approximately a million objects among which we visually inspected the 105 closest neighbors from seven of our brightest, most well-sampled simulations, comprising 89 unique ZTF DR4 sources. Our result illustrates the potential of coherently incorporating domain knowledge and automatic learning algorithms, which is one of the guiding principles directing the SNAD team. It also demonstrates that the ZTF DR is a suitable testing ground for data mining algorithms aiming to prepare for the next generation of astronomical data., Comment: 30 pages, 5 figures, 3 tables, preprint accepted to New Astronomy

We consider a new mechanism for the removal of the angular momentum from an X-ray binary system and the change in its orbital period - the mass loss in the form of a wind from an accretion disc. Both observations and models predict powerful winds from discs in X-ray transients. We have obtained an analytical estimate for the increase in the orbital period of a binary system with a wind from the disc during an outburst, and quantitative estimates are given for the systems XTE J1118+480, A0620-00 and GRS 1124-68. Resulting rate of period grow is of order of the observed rates of secular decrease in the period. We also compare the predicted rate of change in the period of a binary system due to the flow of matter into the disc and outflow from the second Lagrange point with the observations. It is concluded that the above mechanisms cannot explain the observed secular decrease in the period of the three X-ray Novae, and it is necessary to consider a circumbinary disc that drains the binary's angular momentum., Comment: 10 pages, 7 figures, 2 tables, accepted for publication in Astronomy Letters

We present results from applying the SNAD anomaly detection pipeline to the third public data release of the Zwicky Transient Facility (ZTF DR3). The pipeline is composed of 3 stages: feature extraction, search of outliers with machine learning algorithms and anomaly identification with followup by human experts. Our analysis concentrates in three ZTF fields, comprising more than 2.25 million objects. A set of 4 automatic learning algorithms was used to identify 277 outliers, which were subsequently scrutinised by an expert. From these, 188 (68%) were found to be bogus light curves -- including effects from the image subtraction pipeline as well as overlapping between a star and a known asteroid, 66 (24%) were previously reported sources whereas 23 (8%) correspond to non-catalogued objects, with the two latter cases of potential scientific interest (e. g. 1 spectroscopically confirmed RS Canum Venaticorum star, 4 supernovae candidates, 1 red dwarf flare). Moreover, using results from the expert analysis, we were able to identify a simple bi-dimensional relation which can be used to aid filtering potentially bogus light curves in future studies. We provide a complete list of objects with potential scientific application so they can be further scrutinised by the community. These results confirm the importance of combining automatic machine learning algorithms with domain knowledge in the construction of recommendation systems for astronomy. Our code is publicly available at https://github.com/snad-space/zwad, Comment: 31 pages, 28 figures, 4 tables, accepted for publication in MNRAS

Optical and near-infrared photometry, optical spectroscopy, and soft X-ray and UV monitoring of the changing-look active galactic nucleus NGC 2617 show that it continues to have the appearance of a type-1 Seyfert galaxy. An optical light curve for 2010$-$2017 indicates that the change of type probably occurred between 2010 October and 2012 February and was not related to the brightening in 2013. In 2016 and 2017 NGC 2617 brightened again to a level of activity close to that in 2013 April. However, in 2017 from the end of the March to end of July 2017 it was in very low level and starting to change back to a Seyfert 1.8. We find variations in all passbands and in both the intensities and profiles of the broad Balmer lines. A new displaced emission peak has appeared in H$\beta$. X-ray variations are well correlated with UV$-$optical variability and possibly lead by $\sim$ 2$-$3 d. The $K$ band lags the $J$ band by about 21.5 $\pm$ 2.5 d and lags the combined $B + J$ bands by $\sim$ 25 d. $J$ lags $B$ by $\sim$ 3 d. This could be because $J$-band variability arises predominantly from the outer part of the accretion disc, while $K$-band variability is dominated by thermal re-emission by dust. We propose that spectral-type changes are a result of increasing central luminosity causing sublimation of the innermost dust in the hollow bi-conical outflow. We briefly discuss various other possible reasons that might explain the dramatic changes in NGC 2617., Comment: 4 pages, 3 figures, accepted to the OAP

We present the results of 3D-hydrodynamical simulations of accretion flow in the eclipsing dwarf nova V1239 Her in quiescence. The model includes the optical star filling its Roche lobe, a gas stream emanating from the inner Lagrangian point of the binary system, and the accretion disc structure. A cold hydrogen gas stream is initially emitted towards a point-like gravitational centre. A stationary accretion disc is formed in about 15 orbital periods after the beginning of accretion. The model takes into account partial ionization of hydrogen and uses realistic cooling function for hydrogen. The light curve of the system is calculated as the volume emission of optically thin layers along the line of sight up to the optical depth \tau =2/3 calculated using Planck-averaged opacities. The calculated eclipse light curves show good agreement with observations, with the changing shape of pre-eclipse and post-eclipse light curves being explained entirely due to ~ 50% variations in the mass accretion rate through the gas stream., Comment: 8 pages, 8 figures, 1 movie (ancillary file). Accepted for publication in MNRAS

Optical and near-infrared photometry, optical spectroscopy, and soft X-ray and UV monitoring of the changing look active galactic nucleus NGC 2617 show that it continues to have the appearance of a type-1 Seyfert galaxy. An optical light curve for 2010-2016 indicates that the change of type probably occurred between 2010 October and 2012 February and was not related to the brightening in 2013. In 2016 NGC 2617 brightened again to a level of activity close to that in 2013 April. We find variations in all passbands and in both the intensities and profiles of the broad Balmer lines. A new displaced emission peak has appeared in H$\beta$. X-ray variations are well correlated with UV-optical variability and possibly lead by $\sim$ 2-3 d. The $K$ band lags the $J$ band by about 21.5 $\pm$ 2.5 d. and lags the combined $B+J$ filters by $\sim$ 25 d. $J$ lags $B$ by about 3 d. This could be because $J$-band variability arises from the outer part of the accretion disc, while $K$-band variability comes from thermal re-emission by dust. We propose that spectral-type changes are a result of increasing central luminosity causing sublimation of the innermost dust in the hollow biconical outflow. We briefly discuss various other possible reasons that might explain the dramatic changes in NGC 2617., Comment: 9 pages, 6 figures, accepted for publication in MNRAS

We investigate the viscous evolution of the accretion disk in 4U 1543-47, a black hole binary system, during the first 30 days after the peak of the 2002 burst by comparing the observed and theoretical accretion rate evolution $\dot M(t)$. The observed $\dot M(t)$ is obtained from spectral modelling of the archival RXTE/PCA data. Different scenarios of disk decay evolution are possible depending on a degree of self-irradiation of the disk by the emission from its centre. If the self-irradiation, which is parametrized by factor $C_\mathrm{irr}$, had been as high as $\sim 5\times10^{-3}$, then the disk would have been completely ionized up to the tidal radius and the short time of the decay would have required the turbulent parameter $\alpha\sim 3$. We find that the shape of the $\dot M(t)$ curve is much better explained in a model with a shrinking high-viscosity zone. If $C_\mathrm{irr}\approx(2-3)\times 10^{-4}$, the resulting $\alpha$ lie in the interval $0.5-1.5$ for the black hole masses in the range $6-10~\mathrm{M}_\odot$, while the radius of the ionized disk is variable and controlled by irradiation. For very weak irradiation, $C_\mathrm{irr} < 1.5 \times10^{-4}$, the burst decline develops as in normal outbursts of dwarf novae with $\alpha \sim 0.08-0.32$. The optical data indicate that $C_\mathrm{irr}$ in 4U 1543-47 (2002) was not greater than approximately $(3-6)\times10^{-4}$. Generally, modelling of an X-ray nova burst allows one to estimate $\alpha$ that depends on the black hole parameters. We present the public 1-D code Freddi to model the viscous evolution of an accretion disk. Analytic approximations are derived to estimate $\alpha$ in X-ray novae using $\dot M(t)$., Comment: 13 pages, 14 figures. Accepted for publication in MNRAS. Code Freddi is publicly available at http://xray.sai.msu.ru/~malanchev/freddi/ V3 fixes a typo in an equation on page 3, we thank Ankit Mishra for pointing it out to us

The vertical structure of stationary thin accretion discs is calculated from the energy balance equation with heat generation due to microscopic ion viscosity {\eta} and electron heat conductivity {\kappa}, both depending on temperature. In the optically thin discs it is found that for the heat conductivity increasing with temperature, the vertical temperature gradient exceeds the adiabatic value at some height, suggesting convective instability in the upper disc layer. There is a critical Prandtl number, Pr = 4/9, above which a Keplerian disc become fully convective. The vertical density distribution of optically thin laminar accretion discs as found from the hydrostatic equilibrium equation cannot be generally described by a polytrope but in the case of constant viscosity and heat conductivity. In the optically thick discs with radiation heat transfer, the vertical disc structure is found to be convectively stable for both absorption dominated and scattering dominated opacities, unless a very steep dependence of the viscosity coefficient on temperature is assumed. A polytropic-like structure in this case is found for Thomson scattering dominated opacity., Comment: 8 pages, 3 figures, published in MNRAS