Przemek Wozniak

Energetic eruptions leading to a peculiar hydrogen-rich explosion of a massive star

Every supernova so far observed has been considered to be the terminal explosion of a star. Moreover, all supernovae with absorption lines in their spectra show those lines decreasing in velocity over time, as the ejecta expand and thin, revealing slower-moving material that was previously hidden. In addition, every supernova that exhibits the absorption lines of hydrogen has one main light-curve peak, or a plateau in luminosity, lasting approximately 100 days before declining. Here we report observations of iPTF14hls, an event that has spectra identical to a hydrogen-rich core-collapse supernova, but characteristics that differ extensively from those of known supernovae. The light curve has at least five peaks and remains bright for more than 600 days; the absorption lines show little to no decrease in velocity; and the radius of the line-forming region is more than an order of magnitude bigger than the radius of the photosphere derived from the continuum emission. These characteristics are consistent with a shell of several tens of solar masses ejected by the progenitor star at supernova-level energies a few hundred days before a terminal explosion. Another possible eruption was recorded at the same position in 1954. Multiple energetic pre-supernova eruptions are expected to occur in stars of 95 to 130 solar masses, which experience the pulsational pair instability. That model, however, does not account for the continued presence of hydrogen, or the energetics observed here. Another mechanism for the violent ejection of mass in massive stars may be required.

Real-time detection of optical transients with RAPTOR

Fast variability of optical objects is an interesting though poorly explored subject in modern astronomy. Real-time data processing and identification of transient celestial events in the images is very important for such study as it allows rapid follow-up with more sensitive instruments. We discuss an approach which we have developed for the RAPTOR project, a pioneering closed-loop system combining real-time transient detection with rapid follow-up. RAPTORs data processing pipeline is able to identify and localize an optical transient within seconds after the observation. The testing we performed so far have been confirming the effectiveness of our method for the optical transient detection. The software pipeline we have developed for RAPTOR can easily be applied to the data from other experiments.

Distributed control system for rapid astronomical transient detection

The Rapid Telescope for Optical Response (RAPTOR) program consists of a network of robotic telescopes dedicated to the search for fast optical transients. The pilot project is composed of three observatories separated by approximately 38 kilometers located near Los Alamos, New Mexico. Each of these observatories is composed of a telescope, mount, enclosure, and weather station, all operating robotically to perform individual or coordinated transient searches. The telescopes employ rapidly slewing mounts capable of slewing a 250 pound load 180 degrees in under 2 seconds with arcsecond precision. Each telescope consists of wide-field cameras for transient detection and a narrow-field camera with greater resolution and sensitivity. The telescopes work together by employing a closed-loop system for transient detection and follow-up. Using the combined data from simultaneous observations, transient alerts are generated and distributed via the Internet. Each RAPTOR telescope also has the capability of rapidly responding to external transient alerts received over the Internet from a variety of ground-based and satellite sources. Each observatory may be controlled directly, remotely, or robotically while providing state-of-health and observational results to the client and the other RAPTOR observatories. We discuss the design and implementation of the spatially distributed RAPTOR system.

A portable observatory for persistent monitoring of the night sky

We describe the design and operation of a small, transportable, robotic observatory that has been developed at Los Alamos National Laboratory. This small observatory, called RQD2 (Raptor-Q Design 2), is the prototype for nodes in a global network capable of continuous persistent monitoring of the night sky. The observatory employs five wide-field imagers that altogether view about 90% of the sky above 12 degrees elevation with a sensitivity of R=10 magnitude in 10 seconds. Operating robotically, the RQD2 system acquires a nearly full-sky image every 20 seconds, taking more than 10,000 individual images per night. It also runs real-time astrometric and photometric pipelines that provide both a capability to autonomously search for bright astronomical transients and monitor the variability of optical extinction across the full sky. The first RQD2 observatory began operation in March 2009 and is currently operating at the Fenton Hill site located near Los Alamos, NM.We present a detailed description of the RQD2 system and the data taken during the first several months of operation.

SkyDOT (Sky Database for Objects in the Time Domain): a virtual observatory for variability studies at LANL

The mining of Virtual Observatories (VOs) is becoming a powerful new method for discovery in astronomy. Here we report on the development of SkyDOT (Sky Database for Objects in the Time domain), a new Virtual Observatory, which is dedicated to the study of sky variability. The site will confederate a number of massive variability surveys and enable exploration of the time domain in astronomy. We discuss the architecture of the database and the functionality of the user interface. An important aspect of SkyDOT is that it is continuously updated in near real time so that users can access new observations in a timely manner. The site will also utilize high level machine learning tools that will allow sophisticated mining of the archive. Another key feature is the real time data stream provided by RAPTOR (RAPid Telescopes for Optical Response), a new sky monitoring experiment under construction at Los Alamos National Laboratory (LANL).

Integrating temporal and spectral features of astronomical data using wavelet analysis for source classification

Temporal and spectral information extracted from a stream of photons received from astronomical sources is the foundation on which we build understanding of various objects and processes in the Universe. Typically astronomers fit a number of models separately to light curves and spectra to extract relevant features. These features are then used to classify, identify, and understand the nature of the sources. However, these feature extraction methods may not be optimally sensitive to unknown properties of light curves and spectra. One can use the raw light curves and spectra as features to train classifiers, but this typically increases the dimensionality of the problem, often by several orders of magnitude. We overcome this problem by integrating light curves and spectra to create an abstract image and using wavelet analysis to extract important features from the image. Such features incorporate both temporal and spectral properties of the astronomical data. Classification is then performed on those abstract features. In order to demonstrate this technique, we have used gamma-ray burst (GRB) data from the NASAs Swift mission to classify GRBs into high- and low-redshift groups. Reliable selection of high-redshift GRBs is of considerable interest in astrophysics and cosmology.

Automated variability selection in time-domain imaging surveys using sparse representations with learned dictionaries

Exponential growth in data streams and discovery power delivered by modern time-domain imaging surveys creates a pressing need for variability extraction algorithms that are both fully automated and highly reliable. The current state of the art methods based on image differencing are limited by the fact that for every real variable source the algorithm returns a large number of bogus “detections” caused by atmospheric effects and instrumental signatures coupled with imperfect image processing. Here we present a new approach to this problem inspired by recent advances in computer vision and train the machine to learn new features directly from pixel data. The training data set comes from the Palomar Transient Factory survey and consists of small images centered around transient candidates with known real/bogus classification. This set of high-dimensional vectors (~1000 features) is then transformed into a linear representation using the so called dictionary, an overcomplete feature set constructed separately for each class. The data vectors are well approximated with a small number of dictionary elements, i.e. the dictionary representation is sparse. We show how sparse representations can be used to construct informative features for any suitable machine learning classifier. Our top level classifier is based on the random forest algorithm (collections of decision trees) with input data vectors consisting of up to 6 computer vision features and 20 additional context features designed by subject domain experts. Machine-learned features alone provide only an approximate classification with a 20% missed detection rate at a fixed false positive rate of 1%. When automatically extracted features are appended to those constructed by humans, the rate of missed detections is reduced from 8% to about 4% at 1% false positive rate.

What do telescopes, databases and compute clusters have in common?

Linking ground based telescopes with astronomical satellites, and using the emerging field of intelligent agent architectures to provide crucial autonomous decision making in software, we have combined data archives and research class robotic telescopes along with distributed computing nodes to build an ad-hoc peer-to-peer heterogeneous network of resources. The eSTAR Project* uses intelligent agent technologies to carry out resource discovery, submit observation requests and analyze the reduced data returned from a meta-network of robotic telescopes. We present the current operations paradigm of the eSTAR network and describe the direction of in which the project intends to develop over the next several years. We also discuss the challenges facing the project, including the very real sociological one of user acceptance.