Kirk D. Borne

Ultraluminous infrared galaxies: atlas of near-infrared images

A sample of 27 ultraluminous infrared galaxy (ULIRG) systems has been imaged at 1.6 μm using the Hubble Space Telescope (HST) Near Infrared Camera and Multi-Object Spectrometer (NICMOS). These ULIRGs are from a larger sample also imaged with HST in the I band. Images and catalog information for the NICMOS subsample, as well as brief morphological descriptions of each system are presented. Inspection of the infrared images and a comparison with optical images of these systems shows that at least 85% are obviously composed of two or more galaxies involved in a close interaction or merger event, with as many as 93% showing some signs of interaction history. Approximately 37% of the systems show either spectroscopic or morphological characteristics of an active galactic nucleus. The infrared morphologies of these systems are generally less complicated or disturbed than their optical morphologies, indicating that some of the small-scale features seen in optical images are likely due to complicated patterns of dust obscuration, as well as widely distributed star formation activity. In some systems the high-resolution HST infrared images have revealed nuclear remnants that are obscured or unidentified in ground-based imaging, which has led to changes in previously determined interaction stage classifications or system content. In general, however, the NICMOS images support previous conclusions from previous HST optical imaging.

ESCIENCE AND ARCHIVING FOR SPACE SCIENCE

A confluence of technologies is leading towards revolutionary new interactions between robust data sets, state-of-the-art models and simulations, high-data-rate sensors, and high-performance computing. Data and data systems are central to these new developments in various forms of eScience or grid systems. Space science missions are developing multi-spacecraft, distributed, communications- and computation-intensive, adaptive mission architectures that will further add to the data avalanche. Fortunately, Knowledge Discovery in Database (KDD) tools are rapidly expanding to meet the need for more efficient information extraction and knowledge generation in this data-intensive environment. Concurrently, scientific data management is being augmented by content-based metadata and semantic services. Archiving, eScience and KDD all require a solid foundation in interoperability and systems architecture. These concepts are illustrated through examples of space science data preservation, archiving, and access, including application of the ISO-standard Open Archive Information System (OAIS) architecture.

Scalable, asynchronous, distributed eigen monitoring of astronomy data streams

In this paper, we develop a distributed algorithm for monitoring the principal components (PCs) for next generation of astronomy petascale data pipelines such as the Large Synoptic Survey Telescopes (LSST). This telescope will take repeated images of the night sky every 20 s, thereby generating 30 terabytes of calibrated imagery every night that will need to be co-analyzed with other astronomical data stored at different locations around the world. Event detection, classification, and isolation in such data sets may provide useful insights to unique astronomical phenomenon displaying astrophysically significant variations: quasars, supernovae, variable stars, and potentially hazardous asteroids. However, performing such data mining tasks is a challenging problem for such high-throughput distributed data streams. In this paper, we propose a highly scalable and distributed asynchronous algorithm for monitoring the PCs of such dynamic data streams and discuss a prototype web-based system PADMINI (Peer-to-Peer Astronomy Data Mining) which implements this algorithm for use by the astronomers. We demonstrate the algorithm on a large set of distributed astronomical data to accomplish well-known astronomy tasks such as measuring variations in the fundamental plane of galaxy parameters. The proposed algorithm is provably correct (i.e., converges to the correct PCs without centralizing any data) and can seamlessly handle changes to the data or the network. Real experiments performed on Sloan Digital Sky Survey (SDSS) catalogue data show the effectiveness of the algorithm.

The Shark in the Vitrine: Experiencing our Practice From the Inside Out With Transdisciplinary Lenses

The Scholars of Studying Teaching Collaborative engaged a dozen faculty members from 12 specializations and 4 colleges at a large public university in a 2-year teaching and research project with the goal of learning about and enacting a self-study of professional practice. Participants were selected from various disciplines to provoke alternative perspectives in whole group and critical friend teams. While we each conducted a self-study, we also designed and enacted a meta-study to assess our professional development within the context of the collaborative. We analyze the potential of engaging in collective self-study and report how the methodological challenges initiated transformational learning that bridged theory and praxis. Learning the self-study methodology was complex, but such concentration multiplied the impact of both personal and professional transformation. The study benefits faculty from a broad range of disciplines, at diverse stages in their academic careers, and working at every level of the academic hierarchy.

TagLearner: A P2P Classifier Learning System from Collaboratively Tagged Text Documents

The amount of text data on the Internet is growing at a very fast rate. Online text repositories for news agencies, digital libraries and other organizations currently store giga and tera-bytes of data. Large amounts of unstructured text poses a serious challenge for data mining and knowledge extraction. End user participation coupled with distributed computation can play a crucial role in meeting these challenges. In many applications involving classification of text documents, web users often participate in the tagging process. This collaborative tagging results in the formation of large scale Peer-to-Peer (P2P) systems which can function, scale and self-organize in the presence of highly transient population of nodes and do not need a central server for co-ordination. In this paper, we describe TagLearner, a P2P classifier learning system for extracting patterns from text data where the end users can participate both in the task of labeling the data and building a distributed classifier on it. We present a novel distributed linear programming based classification algorithm which is asynchronous in nature. The paper also provides extensive empirical results on text data obtained from an online repository - the NSF Abstracts Data.

Galaxy Zoo: Mergers – Dynamical models of interacting galaxies

The dynamical history of most merging galaxies is not well understood. Correlations between galaxy interaction and star formation have been found in previous studies, but require the context of the physical history of merging systems for full insight into the processes that lead to enhanced star formation. We present the results of simulations that reconstruct the orbit trajectories and disturbed morphologies of pairs of interacting galaxies. With the use of a restricted three-body simulation code and the help of citizen scientists, we sample 105 points in parameter space for each system. We demonstrate a successful recreation of the morphologies of 62 pairs of interacting galaxies through the review of more than 3 million simulations. We examine the level of convergence and uniqueness of the dynamical properties of each system. These simulations represent the largest collection of models of interacting galaxies to date, providing a valuable resource for the investigation of mergers. This paper presents the simulation parameters generated by the project. They are now publicly available in electronic format at http://data.galaxyzoo.org/mergers.html. Though our best-fitting model parameters are not an exact match to previously published models, our method for determining uncertainty measurements will aid future comparisons between models. The dynamical clocks from our models agree with previous results of the time since the onset of star formation from starburst models in interacting systems and suggest that tidally induced star formation is triggered very soon after closest approach.

The New Computational and Data Sciences Undergraduate Program at George Mason University

We describe the new undergraduate science degree program in Computational and Data Sciences (CDS) at George Mason University (Mason), which began offering courses for both major (B.S.) and minor degrees in Spring 2008. The overarching theme and goal of the program are to train the next-generation scientists in the tools and techniques of cyber-enabled science (e-Science) to prepare them to confront the emerging petascale challenges of data-intensive science. The Mason CDS program has a significantly stronger focus on data-oriented approaches to science than do most computational science and engineering programs. The program has been designed specifically to focus both on simulation (Computational Science) and on data-intensive applications (Data Science). New courses include Introduction to Computational & Data Sciences, Scientific Data and Databases, Scientific Data & Information Visualization, Scientific Data Mining, and Scientific Modeling & Simulation. This is an interdisciplinary science program, drawing examples, classroom materials, and student activities from a broad range of physical and biological sciences. We will describe some of the motivations and early results from the program. More information is available at http://cds.gmu.edu/.

The LSST Data Mining Research Agenda

We describe features of the LSST science database that are amenable to scientific data mining, object classification, outlier identification, anomaly detection, image quality assurance, and survey science validation. The data mining research agenda includes: scalability (at petabytes scales) of existing machine learning and data mining algorithms; development of grid‐enabled parallel data mining algorithms; designing a robust system for brokering classifications from the LSST event pipeline (which may produce 10,000 or more event alerts per night); multi‐resolution methods for exploration of petascale databases; indexing of multi‐attribute multi‐dimensional astronomical databases (beyond spatial indexing) for rapid querying of petabyte databases; and more.

Automated Wildfire Detection Through Artificial Neural Networks

Satellite observations of wildland, agricultural and prescribed fires are routinely identified by Fire Analysts and software algorithms which are part of NOAA’s Hazard Mapping System. However, to more fully automate this operational system, NOAA teamed up with NASA to train a neural network which mimicked the behavior of both analysts and software algorithms in their examination of multi-sensor, satellite imagery. Three spectral channels from GOES, AVHRR and MODIS imagery, spanning the 2003 fire season across the continental United States, was used as training data and JOONE, SNNS and MATLAB software packages generated the NNs. Test results from MATLAB’s 147-10-1 feedforward, backpropagation NN are described. Performance analysis consisted of error matrices, Producer’s, User’s and Overall accuracy, and a Kappa/KHAT.

Parametrization and Classification of 20 Billion LSST Objects: Lessons from SDSS

The Large Synoptic Survey Telescope (LSST) will be a large, wide‐field ground‐based system designed to obtain, starting in 2015, multiple images of the sky that is visible from Cerro Pachon in Northern Chile. About 90% of the observing time will be devoted to a deep‐wide‐fast survey mode which will observe a 20,000 deg2 region about 1000 times during the anticipated 10 years of operations (distributed over six bands, ugrizy). Each 30‐second long visit will deliver 5σ depth for point sources of r∼24.5 on average. The co‐added map will be about 3 magnitudes deeper, and will include 10 billion galaxies and a similar number of stars. We discuss various measurements that will be automatically performed for these 20 billion sources, and how they can be used for classification and determination of source physical and other properties. We provide a few classification examples based on SDSS data, such as color classification of stars, color‐spatial proximity search for wide‐angle binary stars, orbital‐color classi...