Transient Candidates Research Articles

Transient-optimized real-bogus classification with Bayesian convolutional neural networks – sifting the GOTO candidate stream

ABSTRACT Large-scale sky surveys have played a transformative role in our understanding of astrophysical transients, only made possible by increasingly powerful machine learning-based filtering to accurately sift through the vast quantities of incoming data generated. In this paper, we present a new real-bogus classifier based on a Bayesian convolutional neural network that provides nuanced, uncertainty-aware classification of transient candidates in difference imaging, and demonstrate its application to the datastream from the GOTO wide-field optical survey. Not only are candidates assigned a well-calibrated probability of being real, but also an associated confidence that can be used to prioritize human vetting efforts and inform future model optimization via active learning. To fully realize the potential of this architecture, we present a fully automated training set generation method which requires no human labelling, incorporating a novel data-driven augmentation method to significantly improve the recovery of faint and nuclear transient sources. We achieve competitive classification accuracy (FPR and FNR both below 1 per cent) compared against classifiers trained with fully human-labelled data sets, while being significantly quicker and less labour-intensive to build. This data-driven approach is uniquely scalable to the upcoming challenges and data needs of next-generation transient surveys. We make our data generation and model training codes available to the community.

The Young Supernova Experiment: Survey Goals, Overview, and Operations

Abstract Time-domain science has undergone a revolution over the past decade, with tens of thousands of new supernovae (SNe) discovered each year. However, several observational domains, including SNe within days or hours of explosion and faint, red transients, are just beginning to be explored. Here we present the Young Supernova Experiment (YSE), a novel optical time-domain survey on the Pan-STARRS telescopes. Our survey is designed to obtain well-sampled griz light curves for thousands of transient events up to z ≈ 0.2. This large sample of transients with four-band light curves will lay the foundation for the Vera C. Rubin Observatory and the Nancy Grace Roman Space Telescope, providing a critical training set in similar filters and a well-calibrated low-redshift anchor of cosmologically useful SNe Ia to benefit dark energy science. As the name suggests, YSE complements and extends other ongoing time-domain surveys by discovering fast-rising SNe within a few hours to days of explosion. YSE is the only current four-band time-domain survey and is able to discover transients as faint as ∼21.5 mag in gri and ∼20.5 mag in z, depths that allow us to probe the earliest epochs of stellar explosions. YSE is currently observing approximately 750 deg2 of sky every 3 days, and we plan to increase the area to 1500 deg2 in the near future. When operating at full capacity, survey simulations show that YSE will find ∼5000 new SNe per year and at least two SNe within 3 days of explosion per month. To date, YSE has discovered or observed 8.3% of the transient candidates reported to the International Astronomical Union in 2020. We present an overview of YSE, including science goals, survey characteristics, and a summary of our transient discoveries to date.

SkyMapper optical follow-up of gravitational wave triggers: Alert science data pipeline and LIGO/Virgo O3 run

Abstract We present an overview of the SkyMapper optical follow-up programme for gravitational-wave event triggers from the LIGO/Virgo observatories, which aims at identifying early GW170817-like kilonovae out to $\sim200\,\mathrm{Mpc}$ distance. We describe our robotic facility for rapid transient follow-up, which can target most of the sky at $\delta<+10\deg $ to a depth of $i_\mathrm{AB}\approx 20\,\mathrm{mag}$ . We have implemented a new software pipeline to receive LIGO/Virgo alerts, schedule observations and examine the incoming real-time data stream for transient candidates. We adopt a real-bogus classifier using ensemble-based machine learning techniques, attaining high completeness ( $\sim98\%$ ) and purity ( $\sim91\%$ ) over our whole magnitude range. Applying further filtering to remove common image artefacts and known sources of transients, such as asteroids and variable stars, reduces the number of candidates by a factor of more than 10. We demonstrate the system performance with data obtained for GW190425, a binary neutron star merger detected during the LIGO/Virgo O3 observing campaign. In time for the LIGO/Virgo O4 run, we will have deeper reference images allowing transient detection to $i_\mathrm{AB}\approx 21\,\mathrm{mag}$ .

Vetting the optical transient candidates detected by the GWAC network using convolutional neural networks

ABSTRACT The observation of the transient sky through a multitude of astrophysical messengers has led to several scientific breakthroughs in the last two decades, thanks to the fast evolution of the observational techniques and strategies employed by the astronomers. Now, it requires to be able to coordinate multiwavelength and multimessenger follow-up campaigns with instruments both in space and on ground jointly capable of scanning a large fraction of the sky with a high-imaging cadency and duty cycle. In the optical domain, the key challenge of the wide field-of-view telescopes covering tens to hundreds of square degrees is to deal with the detection, identification, and classification of hundreds to thousands of optical transient (OT) candidates every night in a reasonable amount of time. In the last decade, new automated tools based on machine learning approaches have been developed to perform those tasks with a low computing time and a high classification efficiency. In this paper, we present an efficient classification method using convolutional neural networks (CNNs) to discard many common types of bogus falsely detected in astrophysical images in the optical domain. We designed this tool to improve the performances of the OT detection pipeline of the Ground Wide field Angle Cameras (GWAC) telescopes, a network of robotic telescopes aiming at monitoring the OT sky down to R = 16 with a 15 s imaging cadency. We applied our trained CNN classifier on a sample of 1472 GWAC OT candidates detected by the real-time detection pipeline.

The Zwicky Transient Facility Bright Transient Survey. I. Spectroscopic Classification and the Redshift Completeness of Local Galaxy Catalogs

Abstract The Zwicky Transient Facility (ZTF) is performing a three-day cadence survey of the visible northern sky (∼3π) with newly found transient candidates announced via public alerts. The ZTF Bright Transient Survey (BTS) is a large spectroscopic campaign to complement the photometric survey. BTS endeavors to spectroscopically classify all extragalactic transients with m peak ≤ 18.5 mag in either the g ZTF or r ZTF filters, and publicly announce said classifications. BTS discoveries are predominantly supernovae (SNe), making this the largest flux-limited SN survey to date. Here we present a catalog of 761 SNe, classified during the first nine months of ZTF (2018 April 1–2018 December 31). We report BTS SN redshifts from SN template matching and spectroscopic host-galaxy redshifts when available. We analyze the redshift completeness of local galaxy catalogs, the redshift completeness fraction (RCF; the ratio of SN host galaxies with known spectroscopic redshift prior to SN discovery to the total number of SN hosts). Of the 512 host galaxies with SNe Ia, 227 had previously known spectroscopic redshifts, yielding an RCF estimate of 44% ± 4%. The RCF decreases with increasing distance and decreasing galaxy luminosity (for z < 0.05, or ∼200 Mpc, RCF ≈ 0.6). Prospects for dramatically increasing the RCF are limited to new multifiber spectroscopic instruments or wide-field narrowband surveys. Existing galaxy redshift catalogs are only ∼50% complete at r ≈ 16.9 mag. Pushing this limit several magnitudes deeper will pay huge dividends when searching for electromagnetic counterparts to gravitational wave events or sources of ultra-high-energy cosmic rays or neutrinos.

A Real-time Automatic Validation System for Optical Transients Detected by GWAC

The ground-based wide-angle camera array generates millions of single frame alerts per night. After the complicated and elaborate filters by multiple methods, a couple of dozens of candidates are still needed to be confirmed by follow-up observations in real-time. In order to free scientists from the complex and high-intensity follow-up tasks, we developed a Real-time Automatic transient Validation System (RAVS), and introduce here its system architecture, data processing flow, database schema, automatic follow-up control flow, and mobile message notification solution. This system is capable of automatically carrying out all operations in real-time without human intervention, including the validation of transient candidates, the adaptive light-curve sampling for identified targets in multi-band, and the pushing of observation results to the mobile client. The running of RAVS shows that an M-type stellar flare event can be well sampled by RAVS without a significant loss of the details, while the observing time is only less than one-third of the time coverage. Because the control logic of RAVS is designed to be independent of the telescope hardware, RAVS can be conveniently transplanted to other telescopes, especially the follow-up system of SVOM. Some future improvements are presented for the adaptive light-curve sampling, after taking into account both the brightness of sources and the evolution trends of the corresponding light-curves.

Palomar Gattini-IR: Survey Overview, Data Processing System, On-sky Performance and First Results

Palomar Gattini-IR is a new wide-field, near-infrared (NIR) robotic time domain survey operating at Palomar Observatory. Using a 30 cm telescope mounted with a H2RG detector, Gattini-IR achieves a field of view (FOV) of 25 sq. deg. with a pixel scale of 8.″7 in J-band. Here, we describe the system design, survey operations, data processing system and on-sky performance of Palomar Gattini-IR. As a part of the nominal survey, Gattini-IR scans ≈7500 square degrees of the sky every night to a median 5σ depth of 15.7 AB mag outside the Galactic plane. The survey covers ≈15,000 square degrees of the sky visible from Palomar with a median cadence of 2 days. A real-time data processing system produces stacked science images from dithered raw images taken on sky, together with point-spread function (PSF)-fit source catalogs and transient candidates identified from subtractions within a median delay of ≈4 hr from the time of observation. The calibrated data products achieve an astrometric accuracy (rms) of ≈0.″7 with respect to Gaia DR2 for sources with signal-to-noise ratio > 10, and better than ≈0.″35 for sources brighter than ≈12 Vega mag. The photometric accuracy (rms) achieved in the PSF-fit source catalogs is better than ≈3% for sources brighter than ≈12 Vega mag and fainter than the saturation magnitude of ≈8.5 Vega mag, as calibrated against the Two Micron All Sky Survey catalog. The detection efficiency of transient candidates injected into the images is better than 90% for sources brighter than the 5σ limiting magnitude. The photometric recovery precision of injected sources is 3% for sources brighter than 13 mag, and the astrometric recovery rms is ≈0.″9. Reference images generated by stacking several field visits achieve depths of ≳16.5 AB mag over 60% of the sky, while it is limited by confusion in the Galactic plane. With a FOV ≈40× larger than any other existing NIR imaging instrument, Gattini-IR is probing the reddest and dustiest transients in the local universe such as dust obscured supernovae in nearby galaxies, novae behind large columns of extinction within the galaxy, reddened microlensing events in the Galactic plane and variability from cool and dust obscured stars. We present results from transients and variables identified since the start of the commissioning period.

The mini-GWAC optical follow-up of gravitational wave alerts – results from the O2 campaign and prospects for the upcoming O3 run

The second (O2) observational campaign of gravitational waves (GWs) organized by the LIGO/Virgo Collaborations has led to several breakthroughs such as the detection of GW signals from merger systems involving black holes or neutrons stars. During O2, 14 GW alerts were sent to the astronomical community with sky regions mostly covering over hundreds of square degrees. Among them, six were finally confirmed as real astrophysical events. Since 2013, a new set of ground-based robotic telescopes called Ground-based Wide Angle Camera system (GWAC) project and its pathfinder mini-GWAC has been developed to contribute to the various challenges of multi-messenger and time domain astronomy. The GWAC system is built up in the framework of the ground-segment system of the SVOM mission that will be devoted to the study of the multi-wavelength transient sky in the next decade. During O2, only the mini-GWAC telescope network was fully operational. Due to the wide field of view and fast automatic follow-up capabilities of the mini-GWAC telescopes, they were adept to efficiently cover the sky localization areas of GW event candidates. In this paper, we present the mini-GWAC pipeline we have set up to respond to GW alerts and we report our optical follow-up observations of eight GW alerts detected during the O2 run. Our observations provided the largest coverage of the GW localization areas with a short latency made by any optical facility. We found tens of optical transient candidates in our images, but none of those could be securely associated with any confirmed black hole – black hole merger event. Based on this first experience and the near future technical improvements of our network system, we will be more competitive in detecting the optical counterparts from some GW events that will be identified during the upcoming O3 run, especially those emerging from binary neutron star mergers.

Searching for fast extragalactic X-ray transients in Chandra surveys

ABSTRACT Recent works have discovered two fast (≈10 ks) extragalactic X-ray transients in the Chandra Deep Field-South (CDF-S XT1 and XT2). These findings suggest that a large population of similar extragalactic transients might exist in archival X-ray observations. We develop a method that can effectively detect such transients in a single Chandra exposure, and systematically apply it to Chandra surveys of CDF-S, CDF-N, DEEP2, UDS, COSMOS, and E-CDF-S, totaling 19 Ms of exposure. We find 13 transient candidates, including CDF-S XT1 and XT2. With the aid of available excellent multiwavelength observations, we identify the physical nature of all these candidates. Aside from CDF-S XT1 and XT2, the other 11 sources are all stellar objects, and all of them have z-band magnitudes brighter than 20. We estimate an event rate of ${59^{+77}_{-38}\ \rm {evt\ yr^{-1}\, deg^{-2}}}$ for CDF-S XT-like transients with 0.5–7 keV peak fluxes log Fpeak ≳ −12.6 (erg cm−2 s−1). This event rate translates to ${\approx 15^{+20}_{-10}}$ transients existing among Chandra archival observations at Galactic latitudes |b| > 20°, which can be probed in future work. Future missions such as Athena and the Einstein Probe with large grasps (effective area × field of view) are needed to discover a large sample (∼thousands) of fast extragalactic X-ray transients.

Machine learning on difference image analysis: A comparison of methods for transient detection

We present a comparison of several Difference Image Analysis (DIA) techniques, in combination with Machine Learning (ML) algorithms, applied to the identification of optical transients associated with gravitational wave events. Each technique is assessed based on the scoring metrics of Precision, Recall, and their harmonic mean F1, measured on the DIA results as standalone techniques, and also in the results after the application of ML algorithms, on transient source injections over simulated and real data. This simulations cover a wide range of instrumental configurations, as well as a variety of scenarios of observation conditions, by exploring a multi dimensional set of relevant parameters, allowing us to extract general conclusions related to the identification of transient astrophysical events. The newest subtraction techniques, and particularly the methodology published in Zackay et al. (2016) are implemented in an Open Source Python package, named properimage, suitable for many other astronomical image analyses. This together with the ML libraries we describe, provides an effective transient detection software pipeline. Here we study the effects of the different ML techniques, and the relative feature importances for classification of transient candidates, and propose an optimal combined strategy. This constitutes the basic elements of pipelines that could be applied in searches of electromagnetic counterparts to GW sources.

X-ray properties of two transient ULX candidates in galaxy NGC 7090

Abstract We report the X-ray data analysis of two transient ultraluminous X-ray sources (ULXs; hereafter X1 and X2) located in the nearby galaxy NGC 7090. While they were not detected in the 2004 XMM–Newton and 2005 Chandra observations, their 0.3–10 keV X-ray luminosities reached >3 × 1039 erg s−1in later XMM–Newton or Swift observations, showing increases in flux by a factor of >80 and >300 for X1 and X2, respectively. X1 showed indications of spectral variability: at the highest luminosity, its X-ray spectra can be fitted with a power law (Γ = 1.55 ± 0.15), or a multicolour disc model with $T_{\mathrm{in}}=2.07^{+0.30}_{-0.23}$ keV; the X-ray spectrum became softer ($\Gamma =2.67^{+0.69}_{-0.64}$), or cooler ($T_\mathrm{in}=0.64^{+0.28}_{-0.17}$ keV) at lower luminosity. No strong evidence for spectral variability was found for X2. Its X-ray spectra can be fitted with a simple power-law model ($\Gamma =1.61^{+0.55}_{-0.50}$), or a multicolour disc model ($1.69^{+1.17}_{-0.48}$ keV). A possible optical counterpart for X1 is revealed in HST imaging. No optical variability is found, indicating that the optical radiation may be dominated by the companion star. Future X-ray and optical observations are necessary to determine the true nature of the compact object.

Limits on radio emission from meteors using the MWA

Recently, low frequency, broadband radio emission has been observed accompanying bright meteors by the Long Wavelength Array (LWA). The broadband spectra between 20 and 60 MHz were captured for several events, while the spectral index (dependence of flux density on frequency, with $S_\nu \propto \nu^\alpha$) was estimated to be $-4\pm1$ during the peak of meteor afterglows. Here we present a survey of meteor emission and other transient events using the Murchison Widefield Array (MWA) at 72-103 MHz. In our 322-hour survey, down to a $5\sigma$ detection threshold of 3.5 Jy/beam, no transient candidates were identified as intrinsic emission from meteors. We derived an upper limit of -3.7 (95% confidence limit) on the spectral index in our frequency range. We also report detections of other transient events, like reflected FM broadcast signals from small satellites, conclusively demonstrating the ability of the MWA to detect and track space debris on scales as small as 0.1 m in low Earth orbits.

GRAWITA: VLT Survey Telescope observations of the gravitational wave sources GW150914 and GW151226

We report the results of deep optical follow-up surveys of the first two gravitational-wave sources, GW150914 and GW151226, done by the GRAvitational Wave Inaf TeAm Collaboration (GRAWITA). The VLT Survey Telescope (VST) responded promptly to the gravitational-wave alerts sent by the LIGO and Virgo Collaborations, monitoring a region of $90$ deg$^2$ and $72$ deg$^2$ for GW150914 and GW151226, respectively, and repeated the observations over nearly two months. Both surveys reached an average limiting magnitude of about 21 in the $r-$band. The paper describes the VST observational strategy and two independent procedures developed to search for transient counterpart candidates in multi-epoch VST images. Several transients have been discovered but no candidates are recognized to be related to the gravitational-wave events. Interestingly, among many contaminant supernovae, we find a possible correlation between the supernova VSTJ57.77559-59.13990 and GRB150827A detected by {\it Fermi}-GBM. The detection efficiency of VST observations for different types of electromagnetic counterparts of gravitational-wave events are evaluated for the present and future follow-up surveys.

The Electromagnetic Counterpart of the Binary Neutron Star Merger LIGO/Virgo GW170817. I. Discovery of the Optical Counterpart Using the Dark Energy Camera

Abstract We present the Dark Energy Camera (DECam) discovery of the optical counterpart of the first binary neutron star merger detected through gravitational-wave emission, GW170817. Our observations commenced 10.5 hr post-merger, as soon as the localization region became accessible from Chile. We imaged 70 deg2 in the i and z bands, covering 93% of the initial integrated localization probability, to a depth necessary to identify likely optical counterparts (e.g., a kilonova). At 11.4 hr post-merger we detected a bright optical transient located from the nucleus of NGC 4993 at redshift z = 0.0098, consistent (for km s−1 Mpc−1) with the distance of 40 ± 8 Mpc reported by the LIGO Scientific Collaboration and the Virgo Collaboration (LVC). At detection the transient had magnitudes of and , and thus an absolute magnitude of , in the luminosity range expected for a kilonova. We identified 1500 potential transient candidates. Applying simple selection criteria aimed at rejecting background events such as supernovae, we find the transient associated with NGC 4993 as the only remaining plausible counterpart, and reject chance coincidence at the 99.5% confidence level. We therefore conclude that the optical counterpart we have identified near NGC 4993 is associated with GW170817. This discovery ushers in the era of multi-messenger astronomy with gravitational waves and demonstrates the power of DECam to identify the optical counterparts of gravitational-wave sources.

Machine Learning for Pulsar Detection

AbstractThe next generation of radio telescopes will have unprecedented sensitivity and time-resolution offering exciting new capabilities in time-domain science. However, this will result in very large numbers of pulsar and transient event candidates and the associated data rates will be technically challenging in terms of data storage and signal processing. Automated detection and classification techniques are therefore required and must be optimized to allow high-throughput data processing in real time. In this paper we provide a summary of the emerging machine learning techniques being applied to this problem.

Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection∗

Abstract We introduce Deep-HiTS, a rotation-invariant convolutional neural network (CNN) model for classifying images of transient candidates into artifacts or real sources for the High cadence Transient Survey (HiTS). CNNs have the advantage of learning the features automatically from the data while achieving high performance. We compare our CNN model against a feature engineering approach using random forests (RFs). We show that our CNN significantly outperforms the RF model, reducing the error by almost half. Furthermore, for a fixed number of approximately 2000 allowed false transient candidates per night, we are able to reduce the misclassified real transients by approximately one-fifth. To the best of our knowledge, this is the first time CNNs have been used to detect astronomical transient events. Our approach will be very useful when processing images from next generation instruments such as the Large Synoptic Survey Telescope. We have made all our code and data available to the community for the sake of allowing further developments and comparisons at https://github.com/guille-c/Deep-HiTS.

Collaborative Workspaces to Accelerate Discovery

AbstractBy applying a display ecology to the Deeper, Wider, Faster proactive, simultaneous telescope observing campaign, we have shown a dramatic reduction in the time taken to inspect DECam CCD images for potential transient candidates and to produce time-critical triggers to standby telescopes. We also show how facilitating rapid corroboration of potential candidates and the exclusion of non-candidates improves the accuracy of detection; and establish that a practical and enjoyable workspace can improve the experience of an otherwise taxing task for astronomers. We provide a critical road test of two advanced displays in a research context—a rare opportunity to demonstrate how they can be used rather than simply discuss how they might be used to accelerate discovery.

The IPAC Image Subtraction and Discovery Pipeline for the Intermediate Palomar Transient Factory

We describe the near real-time transient-source discovery engine for the intermediate Palomar Transient Factory (iPTF), currently in operations at the Infrared Processing and Analysis Center (IPAC), Caltech. We coin this system the IPAC/iPTF Discovery Engine (or IDE). We review the algorithms used for PSF-matching, image subtraction, detection, photometry, and machine-learned (ML) vetting of extracted transient candidates. We also review the performance of our ML classifier. For a limiting signal-to-noise ratio of 4 in relatively unconfused regions, bogus candidates from processing artifacts and imperfect image subtractions outnumber real transients by ≃10:1. This can be considerably higher for image data with inaccurate astrometric and/or PSF-matching solutions. Despite this occasionally high contamination rate, the ML classifier is able to identify real transients with an efficiency (or completeness) of ≃97% for a maximum tolerable false-positive rate of 1% when classifying raw candidates. All subtraction-image metrics, source features, ML probability-based real-bogus scores, contextual metadata from other surveys, and possible associations with known Solar System objects are stored in a relational database for retrieval by the various science working groups. We review our efforts in mitigating false-positives and our experience in optimizing the overall system in response to the multitude of science projects underway with iPTF.

Intermediate Palomar Transient Factory: Realtime Image Subtraction Pipeline

A fast-turnaround pipeline for realtime data reduction plays an essential role in discovering and permitting follow-up observations to young supernovae and fast-evolving transients in modern time-domain surveys. In this paper, we present the realtime image subtraction pipeline in the intermediate Palomar Transient Factory. By using high-performance computing, efficient databases, and machine-learning algorithms, this pipeline manages to reliably deliver transient candidates within 10 minutes of images being taken. Our experience in using high-performance computing resources to process big data in astronomy serves as a trailblazer to dealing with data from large-scale time-domain facilities in the near future.

IPTF SEARCH FOR AN OPTICAL COUNTERPART TO GRAVITATIONAL-WAVE TRANSIENT GW150914

ABSTRACT The intermediate Palomar Transient Factory (iPTF) autonomously responded to and promptly tiled the error region of the first gravitational-wave event GW150914 to search for an optical counterpart. Only a small fraction of the total localized region was immediately visible in the northern night sky, due both to Sun-angle and elevation constraints. Here, we report on the transient candidates identified and rapid follow-up undertaken to determine the nature of each candidate. Even in the small area imaged of 126 deg2, after extensive filtering, eight candidates were deemed worthy of additional follow-up. Within two hours, all eight were spectroscopically classified by the Keck II telescope. Curiously, even though such events are rare, one of our candidates was a superluminous supernova. We obtained radio data with the Jansky Very Large Array and X-ray follow-up with the Swift satellite for this transient. None of our candidates appear to be associated with the gravitational-wave trigger, which is unsurprising given that GW150914 came from the merger of two stellar-mass black holes. This end-to-end discovery and follow-up campaign bodes well for future searches in this post-detection era of gravitational waves.