Search

The implementation of spectroscopy as a tool in the classification of astrophysical objects is a technique that is becoming increasingly useful for modern astronomy. Currently, a collection of space and ground-based telescopes map the sky every night, exponentially increasing data volume over time. The growing number of datasets pushes astronomers to efficiently classify the spectra of observed objects. In this work, we implement the pre-trained algorithms: INCEPTION V3, RESNET 50 and MNIST, that are based on algorithms of Computer Vision (CV) and Convolutional Neural Networks (CNN). We compare the performance of the classifiers applied on spectra taken from the Sloan Digital Sky Survey (SDSS) in its Data Release DR12 with the Baryon Oscillation Spectroscopic Survey (BOSS) spectrograph with a sample of 300000 spectra. In the learning process, the hyperparameters associated with each model were adjusted, and each model was evaluated with accuracy, precision, and sensitivity metrics. The exercise results show that the best classifier for quasar spectra is RESNET 50 with a performance of more than 60% accuracy. In addition, a low loss rate was obtained in the case of star classification with the INCEPTION V3 model. Furthermore, this investigation confirms that this procedure should be introduced as a regular step in the pipeline of ongoing and upcoming spectroscopy surveys in order to optimize time and quality during the manual classification process carried out by astronomers. [ABSTRACT FROM AUTHOR]

We estimate ([M/H], [ α /M]) for 48 million giants and dwarfs in low-dust extinction regions from the Gaia DR3 XP spectra by using tree-based machine learning models trained on APOGEE DR17 and a metal-poor star sample from Li et al. The root mean square error of our estimation is 0.0890 dex for [M/H] and 0.0436 dex for [ α /M], when we evaluate our models on the test data that are not used in training the models. Because the training data is dominated by giants, our estimation is most reliable for giants. The high-[ α /M] stars and low-[ α /M] stars selected by our ([M/H], [ α /M]) show different kinematical properties for giants and low-temperature dwarfs. We further investigate how our machine learning models extract information on ([M/H], [ α /M]). Intriguingly, we find that our models seem to extract information on [ α /M] from Na D lines (589 nm) and Mg i line (516 nm). This result is understandable given the observed correlation between Na and Mg abundances in the literature. The catalog of ([M/H], [ α /M]) as well as their associated uncertainties is publicly available.

Stellar spectra emulators often rely on large grids and tend to reach a plateau in emulation accuracy, leading to significant systematic errors when inferring stellar properties. Our study explores the use of Transformer models to capture long-range information in spectra, comparing their performance to the Payne emulator (a fully connected multilayer perceptron), an expanded version of The Payne, and a convolutional-based emulator. We tested these models on synthetic spectral grids, evaluating their performance by analyzing emulation residuals and assessing the quality of spectral parameter inference. The newly introduced TransformerPayne emulator outperformed all other tested models, achieving a mean absolute error (MAE) of approximately 0.15% when trained on the full grid. The most significant improvements were observed in grids containing between 1000 and 10,000 spectra, with TransformerPayne showing 2–5 times better performance than the scaled-up version of The Payne. Additionally, TransformerPayne demonstrated superior fine-tuning capabilities, allowing for pretraining on one spectral model grid before transferring to another. This fine-tuning approach enabled up to a 10-fold reduction in training grid size compared to models trained from scratch. Analysis of TransformerPayne's attention maps revealed that they encode interpretable features common across many spectral lines of chosen elements. While scaling up The Payne to a larger network reduced its MAE from 1.2% to 0.3% when trained on the full data set, TransformerPayne consistently achieved the lowest MAE across all tests. The inductive biases of the TransformerPayne emulator enhance accuracy, data efficiency, and interpretability for spectral emulation compared to existing methods.

Magnetic fields are now widely recognized as critical at many scales to galactic dynamics and structure, including multiphase pressure balance, dust processing, and star formation. Using imposed magnetic fields cannot reliably model the interstellar medium's (ISM) dynamical structure nor phase interactions. Dynamos must be modeled. ISM models exist of turbulent magnetic fields using small-scale dynamo (SSD). Others model the large-scale dynamo (LSD) organizing magnetic fields at the scale of the disk or spiral arms. Separately, neither can fully describe the galactic magnetic field dynamics nor topology. We model the LSD and SSD together at a sufficient resolution to use the low explicit Lagrangian resistivity required. The galactic SSD saturates within 20 Myr. We show that the SSD is quite insensitive to the presence of an LSD and is even stronger in the presence of a large-scale shear flow. The LSD grows more slowly in the presence of SSD, saturating after 5 Gyr versus 1–2 Gyr in studies where the SSD is weak or absent. The LSD primarily grows in warm gas in the galactic midplane. Saturation of the LSD occurs due to α -quenching near the midplane as the growing mean-field produces a magnetic α that opposes the kinetic α. The magnetic energy in our models of the LSD shows a slightly sublinear response to increasing resolution, indicating that we are converging toward the physical solution at 1 pc resolution. Clustering supernovae in OB associations increases the growth rates for both the SSD and the LSD, compared to a horizontally uniform supernova distribution. [ABSTRACT FROM AUTHOR]

We use 3D k -means clustering to characterize galaxy substructure in the A2146 cluster of galaxies ( z = 0.2343). This method objectively characterizes the cluster’s substructure using projected position and velocity data for 67 galaxies within a 2.305 Mpc circular region centered on the cluster's optical center. The optimal number of substructures is found to be four. Four distinct substructures with rms velocity typical of galaxy groups or low-mass subclusters, when compared to cosmological simulations of galaxy cluster formation, suggest that A2146 is in the early stages of formation. We utilize this disequilibrium, which is so prevalent in galaxy clusters at all redshifts, to construct a radial mass distribution. Substructures are bound but not virialized. This method is in contrast to previous kinematical analyses, which have assumed virialization, and ignored the ubiquitous clumping of galaxies. The best-fitting radial mass profile is much less centrally concentrated than the well-known Navarro–Frenk–White profile, indicating that the dark-matter-dominated mass distribution is flatter pre-equilibrium, becoming more centrally peaked in equilibrium through the merging of the substructure.

Supernovas (SNs) have been one of the most studied events in astronomy. However, there are still no models capable of describing this phenomenon in a general and accurate way. These models generally seek to describe a single type of supernova, requires multiple parameter's values and often do not distinguish between the different light bands of the same curve. Structured grammatical evolution allows the generation of a model with data and a given basal structure, which can be designed considering the nature of the problem for which we are looking for a model. In this case, with some mathematical assumptions we can generate a symbolic regression to obtain a model for different types of SNs and for each light band. We can also use this algorithm to fit the parametric model of the supernova and obtain the value of the variables needed to model it. [ABSTRACT FROM AUTHOR]

The Solar system Notification Alert Processing System ( snaps ) is a Zwicky Transient Facility (ZTF) and Rubin Observatory alert broker that will send alerts to the community regarding interesting events in the solar system. snaps is actively monitoring solar system objects and one of its functions is to compare objects (primarily main belt asteroids) to one another to find those that are outliers relative to the population. In this paper, we use the SNAPShot1 data set, which contains 31,693 objects from ZTF, and derive outlier scores for each of these objects. snaps employs an unsupervised approach; consequently, to derive outlier rankings for each object, we propose four different outlier metrics such that we can explore variants of the outlier scores and add confidence to the outlier rankings. We also provide outlier scores for each object in each permutation of 15 feature spaces, between two and 15 features, which yields 32,752 total feature spaces. We show that we can derive population outlier rankings each month at Rubin Observatory scale using four Nvidia A100 GPUs, and present several avenues of scientific investigation that can be explored using population outlier detection.

Deriving high-quality light curves for asteroids and other periodic sources from survey data is challenging owing to many factors, including the sparsely sampled observational record and diurnal aliasing, which is a signature imparted into the periodic signal of a source that is a function of the observing schedule of ground-based telescopes. In this paper we examine the utility of combining asteroid observational records from the Zwicky Transient Facility and the Transiting Exoplanet Survey Satellite, which are the ground- and space-based facilities, respectively, to determine to what degree the data from the space-based facility can suppress diurnal aliases. Furthermore, we examine several optimizations that are used to derive the rotation periods of asteroids, which we then compare to the reported rotation periods in the literature. Through this analysis we find that we can reliably derive the rotation periods for ∼85% of our sample of 222 objects that are also reported in the literature and that the remaining ∼15% are difficult to reliably derive, as many are asteroids that are insufficiently elongated, which produces a light curve with an insufficient amplitude and, consequently, an incorrect rotation period. We also investigate a binary classification method that biases against reporting incorrect rotation periods. We conclude the paper by assessing the utility of using other ground- or space-based facilities as companion telescopes to the forthcoming Rubin Observatory.

In recent years, automatic classifiers of image cutouts (also called “stamps”) have been shown to be key for fast supernova discovery. The Vera C. Rubin Observatory will distribute about ten million alerts with their respective stamps each night, enabling the discovery of approximately one million supernovae each year. A growing source of confusion for these classifiers is the presence of satellite glints, sequences of point-like sources produced by rotating satellites or debris. The currently planned Rubin stamps will have a size smaller than the typical separation between these point sources. Thus, a larger field-of-view stamp could enable the automatic identification of these sources. However, the distribution of larger stamps would be limited by network bandwidth restrictions. We evaluate the impact of using image stamps of different angular sizes and resolutions for the fast classification of events (active galactic nuclei, asteroids, bogus, satellites, supernovae, and variable stars), using data from the Zwicky Transient Facility. We compare four scenarios: three with the same number of pixels (small field of view with high resolution, large field of view with low resolution, and a multiscale proposal) and a scenario with the full stamp that has a larger field of view and higher resolution. Compared to small field-of-view stamps, our multiscale strategy reduces misclassifications of satellites as asteroids or supernovae, performing on par with high-resolution stamps that are 15 times heavier. We encourage Rubin and its Science Collaborations to consider the benefits of implementing multiscale stamps as a possible update to the alert specification.

R -process enhanced stars with [Eu/Fe] ≥ +0.7 (so-called r -II stars) are believed to have formed in an extremely neutron-rich environment in which a rare astrophysical event (e.g., a neutron-star merger) occurred. This scenario is supported by the existence of an ultra-faint dwarf galaxy, Reticulum II, where most of the stars are highly enhanced in r -process elements. In this scenario, some small fraction of dwarf galaxies around the Milky Way were r enhanced. When each r-enhanced dwarf galaxy accreted to the Milky Way, it deposited many r -II stars in the Galactic halo with similar orbital actions. To search for the remnants of the r -enhanced systems, we analyzed the distribution of the orbital actions of N = 161 r -II stars in the solar neighborhood by using Gaia EDR3 data. Since the observational uncertainty is not negligible, we applied a newly developed greedy optimistic clustering method to the orbital actions of our sample stars. We found six clusters of r -II stars that have similar orbits and chemistry, one of which is a new discovery. Given the apparent phase-mixed orbits of the member stars, we interpret that these clusters are good candidates for remnants of completely disrupted r -enhanced dwarf galaxies that merged with the ancient Milky Way.

We map the surface of Pluto using an unsupervised machine-learning technique using the near-infrared observations of the LEISA/Ralph instrument on board NASA’s New Horizons spacecraft. The principal-component-reduced Gaussian mixture model was implemented to investigate the geographic distribution of the surface units across the dwarf planet. We also present the likelihood of each surface unit at the image pixel level. Average I/F spectra of each unit were analyzed—in terms of the position and strengths of absorption bands of abundant volatiles such as N _2 , CH _4 , and CO and nonvolatile H _2 O—to connect the unit to surface composition, geology, and geographic location. The distribution of surface units shows a latitudinal pattern with distinct surface compositions of volatiles—consistent with the existing literature. However, previous mapping efforts were based primarily on compositional analysis using spectral indices (indicators) or implementation of complex radiative transfer models, which need (prior) expert knowledge, label data, or optical constants of representative end-members. We prove that an application of unsupervised learning in this instance renders a satisfactory result in mapping the spatial distribution of ice compositions without any prior information or label data. Thus, such an application is specifically advantageous for a planetary surface mapping when label data are poorly constrained or completely unknown, because an understanding of surface material distribution is vital for volatile transport modeling at the planetary scale. We emphasize that the unsupervised learning used in this study has wide applicability and can be expanded to other planetary bodies of the solar system for mapping surface material distribution.

Many research fields are now faced with huge volumes of data automatically generated by specialised equipment. Astronomy is a discipline that deals with large collections of images difficult to handle by experts alone. As a consequence, astronomers have been relying on the power of the crowds, as a form of citizen science, for the classification of galaxy images by amateur people. However, the new generation of telescopes that will produce images at a higher rate highlights the limitations of this approach, and the use of machine learning methods for automatic classification is considered essential. The goal of this paper is to shed light on the automated classification of galaxy images exploring two distinct machine learning strategies. First, following the classical approach consisting of feature extraction together with a classifier, we compare the state-of-the-art feature extractor for this problem, the WND-CHARM, with our proposal based on autoencoders for feature extraction on galaxy images. We then compare these results with an end-to-end classification using convolutional neural networks. To better leverage the available citizen science data, we also investigate a pre-training scheme that exploits both amateur- and expert-labelled data. Our experiments reveal that autoencoders greatly speed up feature extraction in comparison with WND-CHARM and both classification strategies, either using convolutional neural networks or feature extraction, reach comparable accuracy. The use of pre-training in convolutional neural networks, however, has allowed us to provide even better results.

The importance of the current role of data-driven science is constantly increasing within Astrophysics, due to the huge amount of multi-wavelength data collected every day, characterized by complex and high-volume information requiring efficient and, as much as possible, automated exploration tools. Furthermore, to accomplish main and legacy science objectives of future or incoming large and deep survey projects, such as James Webb Space Telescope (JWST), James Webb Space Telescope (LSST), and Euclid, a crucial role is played by an accurate estimation of photometric redshifts, whose knowledge would permit the detection and analysis of extended and peculiar sources by disentangling low-z from high-z sources and would contribute to solve the modern cosmological discrepancies. The recent photometric redshift data challenges, organized within several survey projects, like LSST and Euclid, pushed the exploitation of the observed multi-wavelength and multi-dimensional data or ad hoc simulated data to improve and optimize the photometric redshifts prediction and statistical characterization based on both Spectral Energy Distribution (SED) template fitting and machine learning methodologies. They also provided a new impetus in the investigation of hybrid and deep learning techniques, aimed at conjugating the positive peculiarities of different methodologies, thus optimizing the estimation accuracy and maximizing the photometric range coverage, which are particularly important in the high-z regime, where the spectroscopic ground truth is poorly available. In such a context, we summarize what was learned and proposed in more than a decade of research.

In astronomical surveys, such as the Zwicky Transient Facility, supernovae (SNe) are relatively uncommon objects compared to other classes of variable events. Along with this scarcity, the processing of multiband light curves is a challenging task due to the highly irregular cadence, long time gaps, missing values, few observations, etc. These issues are particularly detrimental to the analysis of transient events: SN-like light curves. We offer three main contributions: (1) Based on temporal modulation and attention mechanisms, we propose a deep attention model (TimeModAttn) to classify multiband light curves of different SN types, avoiding photometric or hand-crafted feature computations, missing-value assumptions, and explicit imputation/interpolation methods. (2) We propose a model for the synthetic generation of SN multiband light curves based on the Supernova Parametric Model, allowing us to increase the number of samples and the diversity of cadence. Thus, the TimeModAttn model is first pretrained using synthetic light curves. Then, a fine-tuning process is performed. The TimeModAttn model outperformed other deep learning models, based on recurrent neural networks, in two scenarios: late-classification and early-classification. Also, the TimeModAttn model outperformed a Balanced Random Forest (BRF) classifier (trained with real data), increasing the balanced- F _1 score from ≈.525 to ≈.596. When training the BRF with synthetic data, this model achieved a similar performance to the TimeModAttn model proposed while still maintaining extra advantages. (3) We conducted interpretability experiments. High attention scores were obtained for observations earlier than and close to the SN brightness peaks. This also correlated with an early highly variability of the learned temporal modulation.

We review Big Data in Astronomy and its role in Astronomy Education. At present all-sky and large-area astronomical surveys and their catalogued data span over the whole range of electromagnetic spectrum, from gamma-ray to radio, as well as most important surveys giving optical images, proper motions, variability and spectroscopic data. Most important astronomical databases and archives are presented as well. They are powerful sources for many-sided efficient research using the Virtual Observatory (VO) environment. It is shown that using and analysis of Big Data accumulated in astronomy lead to many new discoveries. Using these data gives a significant advantage for Astronomy Education due to its attractiveness and due to big interest of young generation to computer science and technologies. The Computer Science itself benefits from data coming from the Universe and a new interdisciplinary science Astroinformatics has been created to manage these data. [ABSTRACT FROM AUTHOR]

Many techniques for the detection of interesting stellar structures in astronomical data sets require full phase-space or color information, which is not always available. The first data release of the GAIA satellite, for example, provided highly accurate positions and magnitudes for more than one billion sources. Therefore the question arises if such structures can also be automatically found without waiting for more detailed information in future data releases. In this contribution we propose and compare two conceptually different strategies to find globular clusters in the GAIA DR1 survey. The first approach is a nearest neighbor retrieval and the second an anomaly detection. Both techniques are able to find most of the known globular clusters within our observation frames consistently, as well as potential candidates for further investigation. Furthermore we address approximation approaches to scale the strategy to larger data. [ABSTRACT FROM AUTHOR]

Abstract We introduce MulensModel, a software package for gravitational microlensing modeling. The package provides a framework for calculating microlensing model magnification curves and goodness-of-fit statistics for microlensing events with single and binary lenses as well as a variety of higher-order effects: extended sources with limb-darkening, annual microlensing parallax, satellite microlensing parallax, and binary lens orbital motion. The software could also be used for analysis of the planned microlensing survey by the NASA flag-ship WFIRST satellite. MulensModel is available at https://github.com/rpoleski/MulensModel/. [ABSTRACT FROM AUTHOR]

Abstract fcmaker is a python module that creates astronomical finding charts for Observing Blocks (OBs) on the p2 web server from the European Southern Observatory (ESO). It provides users with the ability to automate the creation of ESO-compliant finding charts for Service Mode and/or Visitor Mode OBs at the Very Large Telescope (VLT). The design of the fcmaker finding charts, based on an intimate knowledge of VLT observing procedures, is fine-tuned to best support night time operations. As an automated tool, fcmaker also provides observers with the means to independently check visually the observing sequence coded inside an OB. This includes, for example, the signs of telescope and position angle offsets. VLT instruments currently supported by fcmaker include MUSE (WFM-AO, WFM-NOAO, NFM), HAWK-I (AO, NOAO), and X-shooter (full support). The fcmaker code is published on a dedicated Github repository under the GNU General Public License, and is also available via pypi. [ABSTRACT FROM AUTHOR]

Abstract Machine learning algorithms are highly useful for the classification of time series data in astronomy in this era of peta-scale public survey data releases. These methods can facilitate the discovery of new unknown events in most astrophysical areas, as well as improving the analysis of samples of known phenomena. Machine learning algorithms use features extracted from collected data as input predictive variables. A public tool called Feature Analysis for Time Series (FATS) has proved an excellent workhorse for feature extraction, particularly light curve classification for variable objects. In this study, we present a major improvement to FATS, which corrects inconvenient design choices, minor details, and documentation for the re-engineering process. This improvement comprises a new Python package called feets , which is important for future code-refactoring for astronomical software tools. [ABSTRACT FROM AUTHOR]

This paper demonstrates the development of ontology for satellite databases. First, I create a computational ontology for the Union of Concerned Scientists (UCS) Satellite Database (UCSSD for short), called the UCS Satellite Ontology (or UCSSO). Second, in developing UCSSO I show that The Space Situational Awareness Ontology (SSAO) (Rovetto & Kelso 2016)--an existing space domain reference ontology--and related ontology work by the author (Rovetto 2015, 2016) can be used either (i) with a database-specific local ontology such as UCSSO, or (ii) in its stead. In case (i), local ontologies such as UCSSO can reuse SSAO terms, perform term mappings, or extend it. In case (ii), the author's orbital space ontology work, such as the SSAO, is usable by the UCSSD and organizations with other space object catalogs, as a reference ontology suite providing a common semantically-rich domain model. The SSAO, UCSSO, and the broader Orbital Space Environment Domain Ontology project is online at and GitHub. This ontology effort aims, in part, to provide accurate formal representations of the domain for various applications. Ontology engineering has the potential to facilitate the sharing and integration of satellite data from federated databases and sensors for safer spaceflight. [ABSTRACT FROM AUTHOR]

Data processing pipelines represent an important slice of the astronomical software library that include chains of processes that transform raw data into valuable information via data reduction and analysis. In this work we present Corral, a Python framework for astronomical pipeline generation. Corral features a Model-View-Controller design pattern on top of an SQL Relational Database capable of handling: custom data models; processing stages; and communication alerts, and also provides automatic quality and structural metrics based on unit testing. The Model-View-Controller provides concept separation between the user logic and the data models, delivering at the same time multi-processing and distributed computing capabilities. Corral represents an improvement over commonly found data processing pipelines in astronomysince the design pattern eases the programmer from dealing with processing flow and parallelization issues, allowing them to focus on the specific algorithms needed for the successive data transformations and at the same time provides a broad measure of quality over the created pipeline. Corral and working examples of pipelines that use it are available to the community at https://github.com/toros-astro . [ABSTRACT FROM AUTHOR]

La Serena School for Data Science is a multidisciplinary program with six editions so far and a constant format: during 10-14 days, a group of ∼30 students (15 from the US, 15 from Chile and 1-3 from Caribbean countries) and ∼9 faculty gather in La Serena (Chile) to complete an intensive program in Data Science with emphasis in applications to astronomy and bio-sciences. The students attend theoretical and hands-on sessions, and, since early on, they work in multidisciplinary groups with their "mentors" (from the faculty) on real data science problems. The SOC and LOC of the school have developed student selection guidelines to maximize diversity. The program is very successful as proven by the high over-subscription rate (factor 5-8) and the plethora of positive testimony, not only from alumni, but also from current and former faculty that keep in contact with them. [ABSTRACT FROM AUTHOR]

The fields of Astrostatistics and Astroinformatics are vital for dealing with the big data issues now faced by astronomy. Like other disciplines in the big data era, astronomy has many V characteristics. In this paper, we list the different data mining algorithms used in astronomy, along with data mining software and tools related to astronomical applications. We present SDSS, a project often referred to by other astronomical projects, as the most successful sky survey in the history of astronomy and describe the factors influencing its success. We also discuss the success of Astrostatistics and Astroinformatics organizations and the conferences and summer schools on these issues that are held annually. All the above indicates that astronomers and scientists from other areas are ready to face the challenges and opportunities provided by massive data volume.

The phenomenological parameters of eclipsing binary stars, which are the prototypes of the EA, EB and EW systems are determined using the expert complex of computer programs, which realizes the NAV (“New Algol Variable”) algorithm (Andronov 2010, 2012) and its possible modifications are discussed, as well as constrains for estimates of some physical parameters of the systems in a case of photometric observations only, such as the degree of eclipse, ratio of the mean surface brightnesses of the components. The half-duration of the eclipse is 0.0617(7), 0.1092(18) and 0.1015(7) for Algol, Lyrae and W UMa, respectively. The brightness ratio is 6.8±1.0, 4.9±1.0 and 1.15±0.13. These results show that the eclipses have a distinct beginning and end not only in EA (as generally assumed), but also in EB and EW-type systems as well. The algorithm may be applied to classification and study of the newly discovered (or poorly studied) eclipsing variables based on own observations or that obtained using photometric surveys. [ABSTRACT FROM AUTHOR]

In this contribution we report the current stage of the MOLecular Dissociation (MOL-D) database which is a web service within the Serbian virtual observatory (SerVO) and node within Virtual Atomic and Molecular Data Center (VAMDC). MOL-D is an atomic and molecular (A&M) database devoted to the modelling of stellar atmospheres, laboratory plasmas, industrial plasmas etc. The initial stage of development was done at the end of 2014, when the service for data connected with hydrogen and helium molecular ions was done. In the next stage of the development of MOL-D, we include new cross-sections and rate coefficients for processes which involve species such as XH+, where X is atom of some metal. Data are important for the exploring of the interstellar medium as well as for the early Universe chemistry and for the modeling of stellar and solar atmospheres. In this poster, we present our ongoing work and plans for the future. [ABSTRACT FROM PUBLISHER]

The orbital debris problem presents an opportunity for international cooperation toward the mutually beneficial goals of orbital debris prevention, mitigation, remediation, and improved space situational awareness (SSA). Achieving these goals requires sharing orbital debris and other SSA data. Toward this, I present an ontological architecture for the orbital debris and related domains, taking steps in the creation of an orbital debris ontology. The purpose of the ontology is to capture general scientific domain knowledge; formally represent the entities within the domain; form, structure, and standardize (where needed) orbital and SSA terminology; and foster semantic interoperability and data-exchange. In doing so I hope to offer a scientifically accurate ontological representation of the orbital domain; contribute to research in astroinformatics, space ontology, and space data management; and improve spaceflight safety by providing a means to capture and communicate informaiton associated with space debris. [ABSTRACT FROM AUTHOR]

Methods for observing space have changed profoundly in the past few decades. The methods needed to detect and record astronomical objects have shifted from conventional observations in the optical range to more sophisticated methods which permit the detection of not only the shape of an object but also the velocity and frequency of emissions in the millimeter-scale wavelength range and the chemical substances from which they originate. The consolidation of radio astronomy through a range of global-scale projects such as the Very Long Baseline Array (VLBA) and the Atacama Large Millimeter/submillimeter Array (ALMA) reinforces the need to develop better methods of data processing that can automatically detect regions of interest (ROIs) within data cubes (position–position–velocity), index them and facilitate subsequent searches via methods based on queries using spatial coordinates and/or velocity ranges. In this article, we present the development of an automatic system for indexing ROIs in data cubes that is capable of automatically detecting and recording ROIs while reducing the necessary storage space. The system is able to process data cubes containing megabytes of data in fractions of a second without human supervision, thus allowing it to be incorporated into a production line for displaying objects in a virtual observatory. We conducted a set of comprehensive experiments to illustrate how our system works. As a result, an index of 3% of the input size was stored in a spatial database, representing a compression ratio equal to 33:1 over an input of 20.875 GB, achieving an index of 773 MB approximately. On the other hand, a single query can be evaluated over our system in a fraction of second, showing that the indexing step works as a shock-absorber of the computational time involved in data cube processing. The system forms part of the Chilean Virtual Observatory (ChiVO), an initiative which belongs to the International Virtual Observatory Alliance (IVOA) that seeks to provide the capability of content-based searches on data cubes to the astronomical community. [ABSTRACT FROM AUTHOR]

We present qrpca, a fast and scalable QR-decomposition principal component analysis package. The software, written in both R and python languages, makes use of torch for internal matrix computations, and enables GPU acceleration, when available. qrpca provides similar functionalities to prcomp (R) and sklearn (python) packages respectively. A benchmark test shows that qrpca can achieve computational speeds 10–20 × faster for large dimensional matrices than default implementations, and is at least twice as fast for a standard decomposition of spectral data cubes. The qrpca source code is made freely available to the community. [ABSTRACT FROM AUTHOR]